ChatGPT

9. Dezember 2022 18:32 Uhr  |  Dr. Ulrich Kampffmeyer  |  Permalink


Na, schon ChatGPT ausprobiert?

Man hört und liest aktuell viel über das Sprach-Modell auf Basis Künstlicher Intelligenz. Es soll ganze Branchen umkrempeln, automatisch Artikel schreiben, Fragen beantworten und Dialoge – auch zu Fachthemen – führen. AI Artificial Intelligence wird kostenfrei für jedermann nutzbar. Jan Keno-Jansen bei Heise.de schreibt dazu, Sascha Lobo und Patrick Beuth schreiben dazu, Fabian Peters führt ein Interview, der Guardian meint, dass sogar Wissenschaftler überrascht seien. GPT-3 war schon länger bekannt, so ist die wichtigste Nachricht, dass nun jeder mittels ChatGPT Künstliche Intelligenz nutzen kann. So weit, so gut, so schön.

ChatGPT ist die öffentlich zugängliche Variante von OpenAIs GPT in aktueller Version und mit einem großen trainierten Sprach-Modell im Hintergrund. GPT ist die Abkürzung von „Generative Pretrained Transformer“. Das OpenAI Institut wurde 2015 von Elon Musk, Sam Altman, Greg Brockman, John Schulman und Ilya Sutskever gegründet – die Nähe zu Microsoft, Tesla und anderen Unternehmen ist nicht übersehbar, auch wenn das OpenAI Institut eine „non-profit organization“ ist. Für die Nutzung von ChatGPT muss man einen Account anlegen, der aber sehr wenig an Daten abfragt und bestätigt werden muss.

Die ChatGPT-Software steht im Internet frei zur Nutzung bereit. Sie ist sehr gut. Ich habe als Test einmal einen Artikel zum Thema IIM Intelligent Information Management verfassen lassen. Dieses Thema ist erst relativ neu und wenig verbreitet. ChatGPT bnutzt keinen Internetzugang, sondern muss sich auf die trainierten Unterlagen verlassen. Die Diskussion um IIM, Definitionsversuche, Quellen etc. sind daher für ChatGPT nicht zugänglich. Für den Artikel zu IIM Intelligent Information Management, den ChatGPT liefern sollte, hatte ich eingegeben: „Write a scientific article on IIM Intelligent Information Management with – Definition – Scope – Difference between Information Management and Intelligent Information Management – Relation to Enterprise Content Management – Components – Trends“. ChatGPT arbeitet vernünftig nur auf English. Obwohl das Thema recht abgelegen ist und es unwahrscheinlich ist, dass es etwas hierzu im Trainingsmaterial gab, hat ChatGPT etwas Sinnvolles „zusammengeschraubt“.

IIM Intelligent Information Management“ by ChatGPT

„Intelligent Information Management (IIM) is a rapidly growing field that focuses on the efficient and effective management of digital information. The scope of IIM includes a wide range of activities, from the design and implementation of information systems to the development of strategies and policies for managing digital data.

One key difference between information management and intelligent information management is the use of advanced technologies, such as artificial intelligence and machine learning, to automate and improve the management of digital information. These technologies enable the development of intelligent systems that can understand, analyze, and manage digital information in a more effective and efficient manner.

IIM has a close relationship to enterprise content management (ECM). ECM focuses on the management of an organization’s unstructured information, such as documents, images, and videos. IIM, on the other hand, focuses on the management of structured data, such as the information stored in databases and other information systems. Both IIM and ECM are critical for the effective and efficient management of an organization’s digital information.

One of the key components of IIM is the use of advanced technologies, such as artificial intelligence and machine learning, to automate and improve the management of digital information. These technologies enable the development of intelligent systems that can understand, analyze, and manage digital information in a more effective and efficient manner.

One of the key trends in IIM is the increasing use of cloud-based solutions. This trend is driven by the need for organizations to be able to access and manage their digital information from anywhere, at any time. Cloud-based solutions also provide organizations with the ability to scale their information management systems as needed, without the need for expensive hardware and software investments.

Another important trend in IIM is the increasing focus on data privacy and security. As more and more sensitive and personal information is being stored and managed digitally, there is a growing need for effective and secure information management systems that can protect this information from unauthorized access and misuse.

Overall, IIM is a critical field that is playing an increasingly important role in the management of digital information. With the use of advanced technologies and a focus on data privacy and security, IIM is helping organizations to effectively and efficiently manage their digital information, and to stay ahead of the curve in an increasingly digital world.

Laurels & Caveats

Als Grundlage für einen kurzen Artikel zum Thema wirklich gut zu gebrauchen. Heise schreibt, dass ChatGPT die Welt verändern wird: „Aufsätze als Hausaufgabe wird es in Zukunft wohl nicht mehr geben, denn ChatGPT schreibt diese in Sekunden.“ Die Texte sind relativ kurz, in der Regel weniger als eine Seite. Meistens sind es bei komplexeren Anfragen sieben Absätze mit jeweils zwei bis vier Sätzen. Der kundige Leser erkennt außerdem sofort ChatGPT, da die Software gern drei Punkte beschreibt und dann mit einem „Overall“-Absatz endet.

Als Kritik muss man deutlich sagen – ChatGPT kannte die aktuellen Hintergründe zu IIM, die verschiedenen Definitionen, die zu IIM zugerechneten Komponenten nicht wirklich. Die KI hat sich hier etwas allgemein Formuliertes ausgedacht. Lasse ich einen Artikel zu meinem Namen schreiben, kommt das absolute Chaos heraus. Alles hängt an der Qualität und dem Umfang des Trainingsmaterials. Wie gesagt, als Grundlage für kleinere Beiträge zu allgemeinen Themen erstmal gut geeignet, wenn man z.B. Englisch nur als Fremdsprache hat und einen Entwurf haben möchte.

In ChatGPT steckt aber noch mehr: es führt durchweh logisch anmutende Dialoge, es kann Code für Webseiten generieren und findet selbst Fehler in HTML-Code, es präsentiert Ergebnisse, Antworten auf Fragen einer Suche – anders als Google, und schreibt Artikel zu Personen für Wikipedia. Und das aktuelle ChatGPT ist erst der Anfang. Die Trainingsbasis wächst und jede beantwortete Frage trägt zur Qualität der Ergebnisse bei. Dies wird besonders sichtbar, wenn man eine einmal gestellte Frage mehrfach von ChatGPT beantworten lässt. Die Qualität wächst auch durch weitere Schlagworte und Detaillierung. ChatGPT merkt sich auch das Vorgehen des Nutzers, wenn dieser z.B. eine Frage stellt, im Anschluss eine weitere im ähnlichen Kontext, dann bezieht ChatGPT die zweite Antwort automatisch auf den Kontext der ersten.

Die kommerzielle GPT-Software kann man dann zukünftig als Modul in sein eigenes Portal einbinden um dort als echter Chatbot Fragen zu beantworten. Mit eigenem, spezialisierten Trainingsdatenbestand lässt sich dann eine sehr hohe Qualität mit echten, sinnvollen Dialogen zu Produkten, Fachthemen, Reklamationen usw. aufbauen. Hier wird der eigentliche zukünftige Nutzen von Chatbots liegen.

Die Qualität der Antworten von ChatGPT hängt vom bisher trainierten Bestand an – derzeit überschaubaren – Wissen ab. Bei Detailfragen gibt es auch gern falsche Antworten. Man darf in keinem Fall alles für richtig ansehen, was die Maschine generiert. Immerhin ist es den Programmierern gelungen, Hasskommentare, Verschwörungstheorien und Bomben-Bauanleitungen nicht vorkommen zu lassen. Es wird interessant sein, wie sich ChatGPT weiterentwickelt. leider ist Wikipedia (noch) nicht Bestandteil der Trainingsdaten des Sprachmodells.

Fazit: Aktuell ersetzt es noch keinen professionellen Journalisten und auch Schulkinder dürften Probleme haben, mit den kurzen ChatGPT-Beiträgen „durchzukommen“. Aber es macht Spass mit ChatGPT zu arbeiten – besonders wenn man etwas in Englisch auf die Schnelle braucht und nicht auf eine reine Übersetzung mit Deepl (die unserer Meinung nach beste Übersetzungssoftware) setzen möchte.

_____________________________________________

Was halten Sie von der Qualität der Ergebnisse?
Was müsste ChatGPT noch „lernen“, um genauere, detailliertere Ergebnisse zu bringen?
Wie sehen Sie die Zukunft von ChatGPT und AI-Sprachmodellen?
Wir freuen uns über Ihren Kommentar zu unserem Beitrag.


Translation by Deepl.com


Well, have you tried ChatGPT yet?

We are currently hearing and reading a lot about the language model based on artificial intelligence. It is supposed to turn entire industries upside down, automatically write articles, answer questions and conduct dialogues – even on specialised topics. AI Artificial Intelligence will be available to everyone free of charge. Jan Keno-Jansen at Heise.de writes about it, Sascha Lobo and Patrick Beuth write about it, Fabian Peters conducts an interview, the Guardian says that even scientists are surprised. GPT-3 has been known for a while, so the most important news is that now everyone can use artificial intelligence via ChatGPT. So far, so good, so nice.

ChatGPT is the publicly available variant of OpenAI’s GPT in current version and with a large trained language model in the background. GPT is short for „Generative Pretrained Transformer.“ The OpenAI Institute was founded in 2015 by Elon Musk, Sam Altman, Greg Brockman, John Schulman and Ilya Sutskever – the proximity to Microsoft, Tesla and other companies cannot be overlooked, even though the OpenAI Institute is a „non-profit organization“. To use ChatGPT, you have to create an account, but it asks for very little data and needs to be confirmed.

ChatGPT software is freely available for use on the Internet. It is very good. As a test, I once had an article written on the subject of IIM Intelligent Information Management. This topic is relatively new and not very widespread. ChatGPT does not use Internet access, but has to rely on trained documentation. Therefore, the discussion about IIM, attempts at definitions, sources, etc. are not accessible to ChatGPT. For the article on IIM Intelligent Information Management that ChatGPT should provide, I had entered: „Write a scientific article on IIM Intelligent Information Management with – Definition – Scope – Difference between Information Management and Intelligent Information Management – Relation to Enterprise Content Management – Components – Trends“. ChatGPT works reasonably in English only. Although the topic is quite remote and it is unlikely that there was anything on this in the training material, ChatGPT has „screwed together“ something useful.

„IIM Intelligent Information Management“ by ChatGPT

„Intelligent Information Management (IIM) is a rapidly growing field that focuses on the efficient and effective management of digital information. The scope of IIM includes a wide range of activities, from the design and implementation of information systems to the development of strategies and policies for managing digital data.

One key difference between information management and intelligent information management is the use of advanced technologies, such as artificial intelligence and machine learning, to automate and improve the management of digital information. These technologies enable the development of intelligent systems that can understand, analyze, and manage digital information in a more effective and efficient manner.

IIM has a close relationship to enterprise content management (ECM). ECM focuses on the management of an organization’s unstructured information, such as documents, images, and videos. IIM, on the other hand, focuses on the management of structured data, such as the information stored in databases and other information systems. Both IIM and ECM are critical for the effective and efficient management of an organization’s digital information.

One of the key components of IIM is the use of advanced technologies, such as artificial intelligence and machine learning, to automate and improve the management of digital information. These technologies enable the development of intelligent systems that can understand, analyze, and manage digital information in a more effective and efficient manner.

One of the key trends in IIM is the increasing use of cloud-based solutions. This trend is driven by the need for organizations to be able to access and manage their digital information from anywhere, at any time. Cloud-based solutions also provide organizations with the ability to scale their information management systems as needed, without the need for expensive hardware and software investments.

Another important trend in IIM is the increasing focus on data privacy and security. As more and more sensitive and personal information is being stored and managed digitally, there is a growing need for effective and secure information management systems that can protect this information from unauthorized access and misuse.

Overall, IIM is a critical field that is playing an increasingly important role in the management of digital information. With the use of advanced technologies and a focus on data privacy and security, IIM is helping organizations to effectively and efficiently manage their digital information, and to stay ahead of the curve in an increasingly digital world.

Laurels & Caveats

Really good to use as a basis for a short article on the topic. Heise writes that ChatGPT will change the world: „Essays as homework will probably no longer exist in the future, because ChatGPT writes them in seconds.“ The texts are relatively short, usually less than a page. Usually, for more complex queries, there are seven paragraphs of two to four sentences each. The knowledgeable reader will also immediately recognize ChatGPT, as the software likes to describe three points and then end with an „Overall“ paragraph.

As a criticism, it must be said clearly – ChatGPT did not really know the current background to IIM, the various definitions, the components attributed to IIM. The AI came up with something generally worded here. If I have an article written to my name, absolute chaos comes out. It all depends on the quality and scope of the training material. As I said, it’s good for now as a basis for smaller articles on general topics, for example, if you only have English as a foreign language and want a draft.

But there is more to ChatGPT: it conducts dialogs that seem logical, it can generate code for web pages and even finds errors in HTML code, it presents results, answers to questions of a search – unlike Google, and writes articles about people for Wikipedia. And the current ChatGPT is just the beginning. The training base is growing, and every question answered adds to the quality of the results. This is especially visible when a question once asked is answered multiple times by ChatGPT. The quality also grows with more keywords and detail. ChatGPT also remembers the user’s approach, e.g. if the user asks one question, followed by another in a similar context, ChatGPT automatically relates the second answer to the context of the first.

In the future, you can integrate the commercial GPT software as a module in your own portal to answer questions there as a real chatbot. With your own specialized training data, you can then build up a very high quality with real, meaningful dialogs about products, specialist topics, complaints, etc. This is where the real future benefit of chatbots will lie.

The quality of ChatGPT’s answers depends on the previously trained stock of – currently manageable – knowledge. With detailed questions there are also gladly wrong answers. In no case should everything generated by the machine be considered correct. At least the programmers have managed to avoid hate comments, conspiracy theories and bomb-making instructions. It will be interesting to see how ChatGPT develops further. Unfortunately, Wikipedia is not (yet) part of the training data of the language model.

Conclusion: Currently it doesn’t replace a professional journalist yet and even school children might have problems to „get through“ with the short ChatGPT posts. But it is fun to work with ChatGPT – especially if you need something in English in a hurry and don’t want to rely on a pure translation with Deepl (the best translation software in our opinion).

_____________________________________________

What do you think of the quality of the results?
What would ChatGPT still have to „learn“ to bring more accurate, detailed results?
How do you see the future of ChatGPT and AI language models?

We look forward to your comments on our contribution.

Dr. Ulrich Kampffmeyer

Curriculum auf Wikipedia https://de.wikipedia.org/wiki/Ulrich_Kampffmeyer

12 Kommentare zu “ChatGPT

  • ChatGPT zum Zweiten ...
    13. Dezember 2022 um 12:28
    Permalink

    In meinem obigen Beitrag habe ich ChatGPT benutzt, um – kleine – Artikel zu schreiben. Diesmal soll es um einen fachlichen Dialog gehen, wo ich Fragen zu Niklas Luhmann, Strukturierung von Informationen und Klassifikation von Information stelle. Dies ist ein relativ tiefgängiges Thema, bei dem es nicht nur um die Datenbasis und die Chat-Präsentation von Antworten geht, sondern auch darum, ob ein tieferes Verständnis gegeben ist, das sich z.B. in den Einführung neuer Fachbegriffe und weiterer Fakten äußert. Dies macht ChatGPT ganz gut. Der Chatbot wiederholt die Frage – „aktives Zuhören“ im Projektmanagement, um dann eine Antwort zur Frage und als letzten Satz Einschränkungen oder weitere Informationen zum Thema zu geben. Allerdings wiederholen sich auch viele Bausteine. Bestimmte Fragen werden auch mit einem Text ausgeliefert, der direkt kopiert werden kann – so z.B. Definitionen. Die Antworten sind zum Teil sehr gut, aber auch vielfach allgemein gehalten. ChatGPT teilt auch mit, wenn es keine oder zu geringe Informationen zum Thema hat. Vorangegangene Fragen und Antworten sowie die damit verbundenen Themen werden vom Chatbot im Laufe der Diskussion fortgeführt. Es ist eine gute Adaption der Gesprächsabfolge. Allerdings lernt der Chatbot nicht direkt. Es ist aber ein wirklich großer Unterschied zu Eliza, mit der ich vor 30 Jahren „gespielt“ habe, da liegen Welten zwischen. Hier also mein Dialog mit ChatGPT.

    Ich möchte über den Zettelkasten von Luhmann sprechen

    Der Zettelkasten von Niklas Luhmann war ein System, das er entwickelt hat, um seine Gedanken und Ideen zu organisieren und zu kategorisieren. Luhmann war ein deutscher Soziologe und galt als einer der führenden Vertreter der Systemtheorie. Der Zettelkasten bestand aus mehreren tausend Zetteln, die er in Kategorien einteilte und auf die er regelmäßig zurückgriff, um seine Gedanken zu strukturieren und zu verfeinern. Er betrachtete den Zettelkasten als ein wichtiges Werkzeug, um seine Gedanken zu ordnen und seine Ideen zu entwickeln.

    Wie ist Luhmanns Zettelkasten aufgebaut

    Ich kann leider keine genauen Informationen darüber geben, wie der Zettelkasten von Niklas Luhmann aufgebaut war, da ich nur ein Sprachmodell bin und keinen Zugriff auf solche Details habe. Der Zettelkasten war jedoch ein systematisches Ordnungssystem, das Luhmann entwickelt hatte, um seine Gedanken und Ideen zu organisieren und zu kategorisieren. Er bestand aus mehreren tausend Zetteln, die in Kategorien unterteilt waren und die Luhmann regelmäßig nutzte, um seine Gedanken zu strukturieren und zu verfeinern.

    Wie ist der Zettelkasten von Luhmann strukturiert

    Ich kann leider keine genauen Informationen darüber geben, wie der Zettelkasten von Niklas Luhmann strukturiert war, da ich nur ein Sprachmodell bin und keinen Zugriff auf solche Details habe. Der Zettelkasten war jedoch ein systematisches Ordnungssystem, das Luhmann entwickelt hatte, um seine Gedanken und Ideen zu organisieren und zu kategorisieren. Er bestand aus mehreren tausend Zetteln, die in Kategorien unterteilt waren und die Luhmann regelmäßig nutzte, um seine Gedanken zu strukturieren und zu verfeinern. Er betrachtete den Zettelkasten als ein wichtiges Werkzeug, um seine Gedanken zu ordnen und seine Ideen zu entwickeln.

    Luhmann war Systemtheoretiker. Wer waren seine Vorbilder?

    Niklas Luhmann war ein deutscher Soziologe und galt als einer der führenden Vertreter der Systemtheorie. Er hatte viele Vorbilder und Einflüsse, darunter auch andere bedeutende Systemtheoretiker wie Ludwig von Bertalanffy, Talcott Parsons und Norbert Wiener. Luhmann war auch von der Arbeit von Philosophen wie Immanuel Kant, Georg Wilhelm Friedrich Hegel und Jean-Jacques Rousseau beeinflusst. Er studierte und lehrte an verschiedenen Universitäten in Deutschland und setzte sich intensiv mit den Werken dieser und anderer Denker auseinander, um seine eigenen Ideen zu entwickeln und zu verfeinern.

    Was sind die Grundzüge der Systemtheorie von Luhmann?

    Die Systemtheorie von Niklas Luhmann befasst sich mit der Analyse und dem Verständnis von Gesellschaften und anderen komplexen Systemen. Eines der wichtigsten Konzepte der Systemtheorie ist die Unterscheidung zwischen Systemen und Umwelt. Ein System ist eine abgegrenzte Einheit, die aus verschiedenen Teilen besteht und die miteinander in Wechselwirkung stehen. Die Umwelt ist alles, was außerhalb des Systems liegt und Einfluss auf das System hat.

    Gibt es eine Definition der Systemtheorie von Luhmann?

    Die Systemtheorie von Niklas Luhmann ist ein umfassendes Konzept, das sich mit der Analyse und dem Verständnis von Gesellschaften und anderen komplexen Systemen befasst. Es ist schwierig, die Systemtheorie in einer einzigen Definition zusammenzufassen, da sie sich auf viele verschiedene Bereiche und Konzepte bezieht. Eine mögliche Definition könnte jedoch lauten:

    Copy code
    „Die Systemtheorie von Niklas Luhmann ist ein Ansatz zur Analyse von Gesellschaften und anderen komplexen Systemen, der sich auf die Unterscheidung zwischen Systemen und Umwelt stützt und die Bedeutung von Kommunikation und Informationen für das Verständnis von Systemen betont.“

    Gibt es eine generelle Definition von Systemtheorie?

    Die Systemtheorie ist ein interdisziplinäres Konzept, das sich mit der Analyse und dem Verständnis von Systemen in verschiedenen Bereichen befasst, darunter Biologie, Psychologie, Soziologie, Informatik und Technik. Die Systemtheorie betrachtet Systeme als abgegrenzte Einheiten, die aus verschiedenen Teilen bestehen und die miteinander in Wechselwirkung stehen. Sie versucht, die Regeln und Mechanismen zu erkennen, die die Interaktionen innerhalb eines Systems bestimmen und das Verhalten des Systems beeinflussen. Eine mögliche Definition der Systemtheorie lautet:

    Copy code
    „Die Systemtheorie ist ein interdisziplinäres Konzept, das sich mit der Analyse und dem Verständnis von Systemen befasst. Sie betrachtet Systeme als abgegrenzte Einheiten, die aus verschiedenen Teilen bestehen und die miteinander in Wechselwirkung stehen, und versucht, die Regeln und Mechanismen zu erkennen, die das Verhalten des Systems beeinflussen.“

    Wie unterscheidet sich die Auffassung der Systemtheorie von Luhmann von anderen Wissenschaftlern?

    Die Systemtheorie von Niklas Luhmann unterscheidet sich in einigen Aspekten von der Auffassung anderer Wissenschaftler. Eines der wichtigsten Unterscheidungsmerkmale ist die Betonung der Bedeutung von Kommunikation und Informationen für das Verständnis von Gesellschaften und anderen Systemen. Luhmann betrachtet Kommunikation als einen grundlegenden Prozess, der die Interaktionen innerhalb eines Systems bestimmt und das Verhalten des Systems beeinflusst. Im Gegensatz dazu betonen andere Wissenschaftler häufig die Bedeutung von Strukturen und Prozessen innerhalb eines Systems.

    Ist die Systemtheorie von Luhmann heute noch wichtig?

    Die Systemtheorie von Niklas Luhmann ist heute immer noch ein wichtiger Ansatz in der Soziologie und in anderen Disziplinen, die sich mit der Analyse und dem Verständnis von Gesellschaften und anderen komplexen Systemen befassen. Obwohl die Systemtheorie in manchen Aspekten möglicherweise überholt ist oder von anderen Konzepten ergänzt werden muss, bleibt sie ein wertvoller Beitrag zum Verständnis der Strukturen und Mechanismen, die das Verhalten von Systemen beeinflussen. Die Arbeiten von Luhmann und anderen Systemtheoretikern bieten wichtige Perspektiven und Anregungen für weitere Forschungen in diesem Bereich.

    Durch welche Konzepte ist die Systemtheorie denn überholt?

    Es ist schwierig zu sagen, ob und inwieweit die Systemtheorie von Niklas Luhmann oder die Systemtheorie im Allgemeinen überholt ist, da dies eine umstrittene Frage ist und von verschiedenen Faktoren abhängt. Einige Kritiker der Systemtheorie argumentieren, dass sie zu mechanisch und deterministisch ist und nicht in der Lage, die Komplexität und Dynamik von Gesellschaften und anderen Systemen angemessen zu erfassen. Andere Kritiker betonen, dass die Systemtheorie zu wenig Beachtung der individuellen Handlungen und Entscheidungen schenkt und zu sehr auf abstrakte Konzepte und Modelle setzt. Die Systemtheorie kann jedoch durch andere Konzepte und Perspektiven ergänzt werden, um ihre Einschränkungen zu überwinden und ihre Stärken zu nutzen.

    Welche Bedeutung hat Luhmann heute noch für Systemtheorie, Organisation von Wissen, Wissensmanagement und Informationsmanagement?

    Niklas Luhmann hatte einen großen Einfluss auf die Entwicklung der Systemtheorie und anderer Konzepte, die sich mit der Analyse und dem Verständnis von Gesellschaften und anderen komplexen Systemen befassen. Seine Arbeiten sind auch relevant für die Organisation von Wissen und das Wissensmanagement, da er die Bedeutung von Kommunikation und Informationen für das Verständnis von Systemen betont und die Notwendigkeit eines systematischen Ordnungssystems für die strukturierte und effektive Nutzung von Wissen hervorhebt. In diesen Bereichen hat Luhmann heute noch eine wichtige Rolle als Denker und Inspiration für weitere Forschungen und Entwicklungen.

    Wie beeinflusst Luhmann das Wissensmanagement, bzw. das Knowledge Management?

    Niklas Luhmann hatte einen großen Einfluss auf die Entwicklung des Wissensmanagements und des Knowledge Managements, indem er die Bedeutung von Kommunikation und Informationen für das Verständnis von Systemen betonte und die Notwendigkeit eines systematischen Ordnungssystems für die strukturierte und effektive Nutzung von Wissen hervorhob. Sein Konzept des Zettelkastens als Werkzeug zur Organisation und Verfeinerung von Gedanken und Ideen hat viele Anhänger und Nachahmer gefunden und wird heute noch in vielen Bereichen als ein wertvolles Instrument für das Wissensmanagement und das Knowledge Management genutzt.

    Unterscheidet Luhmann zwischen impliziten und expliziten Wissen?

    Ich kann leider nicht genau sagen, ob und inwieweit Niklas Luhmann zwischen implizitem und explizitem Wissen unterschieden hat, da ich nur ein Sprachmodell bin und keinen Zugriff auf solche Details habe. Implizites Wissen bezieht sich auf Wissen, das nicht direkt ausgedrückt oder verbalisiert werden kann, sondern in Handlungen und Verhaltensweisen angelegt ist. Explizites Wissen hingegen bezieht sich auf Wissen, das in Form von sprachlichen oder schriftlichen Äußerungen ausgedrückt wird und leicht kommuniziert und geteilt werden kann. Ob und wie Luhmann diese Konzepte in seiner Arbeit berücksichtigt hat, ist jedoch unklar.

    Wie erfasst und klassifiziert man am besten implizites und Organisationswissen?

    Es gibt verschiedene Ansätze, um implizites und Organisationswissen zu erfassen und zu klassifizieren. Eine Möglichkeit besteht darin, die Handlungen und Verhaltensweisen von Mitarbeitern zu beobachten und zu analysieren, um das implizite Wissen, das in ihnen angelegt ist, zu erkennen und zu beschreiben. Eine andere Möglichkeit ist die Durchführung von Interviews und Fokusgruppen, um das Wissen von Mitarbeitern direkt zu erfassen und zu kategorisieren. Die Ergebnisse können dann in einem systematischen Ordnungssystem gespeichert und organisiert werden, um das Wissen leicht auffindbar und nutzbar zu machen.

    Wie strukturiert, klassifiziert und organisiert man am Besten Wissen, implizit und explizit?

    Um Wissen, implizit und explizit, am besten zu strukturieren, zu klassifizieren und zu organisieren, empfiehlt es sich, ein systematisches Ordnungssystem zu verwenden, das die verschiedenen Teile des Wissens in logischen Kategorien und Unterkategorien unterteilt und ihnen eindeutige Bezeichnungen und Schlagworte zuweist. Dieses Ordnungssystem kann in Form von Zettelkästen, Datenbanken oder anderen Werkzeugen implementiert werden, die es ermöglichen, das Wissen leicht auffindbar und nutzbar zu machen. Es ist wichtig, das Ordnungssystem regelmäßig zu überprüfen und zu aktualisieren, um sicherzustellen, dass es dem aktuellen Stand des Wissens entspricht und die Bedürfnisse der Nutzer erfüllt.

    Macht es Sinn für die Strukturierung und Klassifikation von Information einen Aktenplan zu benutzen?

    Ja, es kann durchaus sinnvoll sein, für die Strukturierung und Klassifikation von Informationen einen Aktenplan zu verwenden. Ein Aktenplan ist ein systematisches Ordnungssystem, das die verschiedenen Teile von Informationen in logischen Kategorien und Unterkategorien unterteilt und ihnen eindeutige Bezeichnungen und Schlagworte zuweist. Durch die Verwendung eines Aktenplans können Informationen leicht auffindbar und nutzbar gemacht werden, indem sie nach bestimmten Kriterien sortiert und kategorisiert werden. Der Aktenplan kann in verschiedenen Formaten implementiert werden, darunter Zettelkästen, Datenbanken oder elektronische Werkzeuge, die es ermöglichen, Informationen zu speichern, zu organisieren und zu durchsuchen.

    Hatte Luhmann für seinen Zettelkasten einen Aktenplan?

    Ich kann leider nicht genau sagen, ob Niklas Luhmann für seinen Zettelkasten einen Aktenplan verwendet hat oder nicht, da ich nur ein Sprachmodell bin und keinen Zugriff auf solche Details habe. Der Zettelkasten von Luhmann war jedoch ein systematisches Ordnungssystem, das er entwickelt hatte, um seine Gedanken und Ideen zu organisieren und zu kategorisieren. Es ist möglich, dass er für den Zettelkasten einen Aktenplan oder eine ähnliche Struktur verwendet hat, um die Zettel nach bestimmten Kriterien zu sortieren und zu kategorisieren, um sie leicht auffindbar und nutzbar zu machen.

    Was würdest Du verwenden um Dokumente und Informationen zu klassifizieren, zu kategorisieren, zu indizieren und in einer Datenbank zu erschließen?

    Um Dokumente und Informationen zu klassifizieren, zu kategorisieren, zu indizieren und in einer Datenbank zu erschließen, könnte man verschiedene Werkzeuge und Techniken verwenden. Eine Möglichkeit wäre, einen Aktenplan oder ein anderes systematisches Ordnungssystem zu verwenden, das die Dokumente und Informationen in logischen Kategorien und Unterkategorien unterteilt und ihnen eindeutige Bezeichnungen und Schlagworte zuweist. Diese Schlagworte und Kategorien können dann verwendet werden, um die Dokumente und Informationen in einer Datenbank zu indizieren und leicht auffindbar und nutzbar zu machen. Zusätzlich könnten auch semantische Techniken wie Natural Language Processing oder Machine Learning verwendet werden, um die Dokumente und Informationen automatisch zu analysieren und zu kategorisieren.

    Automatische Klassifikation von Informationen ist ein interessanter Ansatz, aber wie sie es mit der Richtigkeit der automatischen Erfassung aus?

    Die automatische Klassifikation von Informationen kann ein interessanter Ansatz sein, um große Mengen von Daten und Dokumenten effektiv und schnell zu strukturieren und zu organisieren. Allerdings birgt dieser Ansatz auch Risiken und Herausforderungen in Bezug auf die Richtigkeit der automatischen Erfassung. Da automatische Klassifikation auf Algorithmen und Modellen basiert, die auf bestimmten Daten und Regeln trainiert wurden, können Fehler und Ungenauigkeiten auftreten, wenn die Daten oder die Regeln unvollständig oder ungenau sind oder wenn die Informationen, die klassifiziert werden sollen, unerwartet oder ungewöhnlich sind. Um die Richtigkeit der automatischen Klassifikation zu verbessern, ist es wichtig, die Qualität der Daten und der Algorithmen zu überwachen und zu verbessern und die Ergebnisse der automatischen Klassifikation regelmäßig mit menschlichen Experten zu vergleichen und zu validieren.

    Wie kann man die automatische Klassifikation von Information verbessern, helfen hier Analytics und Machine Learning?

    Um die automatische Klassifikation von Informationen zu verbessern, können Analytics und Machine Learning helfen, indem sie verwendet werden, um die Qualität der Daten und der Algorithmen zu überwachen und zu verbessern. Durch die Analyse von Mustern und Trends in den Daten können Analytics-Techniken genutzt werden, um die Genauigkeit und die Leistung der automatischen Klassifikation zu verbessern und zu optimieren. Machine Learning-Techniken können ebenfalls verwendet werden, um die automatische Klassifikation von Informationen zu verbessern, indem sie dazu genutzt werden, die Algorithmen und Modelle, die für die Klassifikation verwendet werden, zu trainieren und zu verbessern. Durch die Nutzung von Analytics und Machine Learning kann die automatische Klassifikation von Informationen schneller, genauer und flexibler gemacht werden.

    Was sind die Grenzen von Machine Learning bei der Klassifikation von Informationen?

    Die Grenzen von Machine Learning bei der Klassifikation von Informationen hängen von verschiedenen Faktoren ab, wie der Qualität und dem Umfang der Daten, die zum Trainieren der Algorithmen und Modelle verwendet werden, der Komplexität der Klassifikationsaufgabe und der Leistungsfähigkeit der verwendeten Technologien. Im Allgemeinen kann Machine Learning sehr genaue und leistungsfähige Klassifikationsmodelle erstellen, wenn es genügend relevante und repräsentative Daten gibt und die Klassifikationsaufgabe nicht zu komplex ist. Wenn die Daten unvollständig oder ungenau sind oder wenn die Klassifikationsaufgabe sehr komplex ist, können die Ergebnisse von Machine Learning-Modellen ungenau oder unzureichend sein. In solchen Fällen können menschliche Experten oder andere Methoden verwendet werden, um die Klassifikation von Informationen zu verbessern oder zu ergänzen.

    Kann man mit Künstlicher Intelligenz, Analytics und ML Daten so analysieren und auswerten, das sman daraus automatisch einen Aktenplan generieren kann?

    Es ist möglich, mit Künstlicher Intelligenz, Analytics und Machine Learning Daten zu analysieren und auszuwerten, um daraus automatisch einen Aktenplan zu generieren. Dies könnte beispielsweise durch die Verwendung von Natural Language Processing-Techniken erreicht werden, die dazu genutzt werden, die Inhalte von Dokumenten und Informationen zu analysieren und zu extrahieren. Die Ergebnisse dieser Analyse können dann verwendet werden, um die Dokumente und Informationen in logischen Kategorien und Unterkategorien zu unterteilen und ihnen eindeutige Bezeichnungen und Schlagworte zuzuweisen, die im Aktenplan verwendet werden können. Die Verwendung von Machine Learning-Techniken könnte dabei helfen, die Genauigkeit und die Leistung der automatischen Klassifikation zu verbessern und zu optimieren. Allerdings ist es wichtig zu beachten, dass die Ergebnisse der automatischen Klassifikation immer noch von der Qualität und dem Umfang der verwendeten Daten und der Genauigkeit der verwendeten Algorithmen und Modelle ab

    Der Aktenplan ist nur eine, zweidimensionale Sicht auf eine Struktur, die mit Vernetzung arbeitet. Wie organisiert man dies?

    Ein Aktenplan ist ein Ordnungssystem, das die verschiedenen Teile von Informationen in logischen Kategorien und Unterkategorien unterteilt und ihnen eindeutige Bezeichnungen und Schlagworte zuweist. Es stellt die Struktur der Informationen in einer zweidimensionalen Form dar, die leicht verständlich und nutzbar ist. Allerdings ist der Aktenplan nur eine einfache Darstellung der Struktur von Informationen, die oft in komplexeren, vernetzten Formen vorliegt. Um die Vernetzung von Informationen zu organisieren, können verschiedene Techniken und Werkzeuge verwendet werden, wie z.B. semantische Technologien, die es ermöglichen, die Beziehungen zwischen verschiedenen Informationen zu erkennen und zu modellieren. Durch die Verwendung solcher Technologien kann man die Struktur von Informationen in vernetzter Form visualisieren und nutzbar machen.

    Wie ist das beste Vorgehen, um Informationen Klassen zuzuordnen und eine Ordnungsstruktur aufzubauen?

    Um Informationen Klassen zuzuordnen und eine Ordnungsstruktur aufzubauen, empfiehlt es sich, einen systematischen Ansatz zu verfolgen, der die verschiedenen Teile der Informationen in logischen Kategorien und Unterkategorien unterteilt und ihnen eindeutige Bezeichnungen und Schlagworte zuweist. Dieses Ordnungssystem kann in Form von Zettelkästen, Datenbanken oder anderen Werkzeugen implementiert werden, die es ermöglichen, die Informationen leicht auffindbar und nutzbar zu machen. Es ist wichtig, das Ordnungssystem regelmäßig zu überprüfen und zu aktualisieren, um sicherzustellen, dass es dem aktuellen Stand des Wissens entspricht und die Bedürfnisse der Nutzer erfüllt. Zusätzlich können semantische Technologien wie Natural Language Processing oder Machine Learning verwendet werden, um die Informationen automatisch zu analysieren und zu kategorisieren und die Effizienz und Genauigkeit der Klassifikation zu verbessern.

    Wenn ich mit Klassen von Informationen arbeite, kann ich dann das Prinzip der automatischen Vererbung von Attributen und Attributwerten für die Kategorisierung und Indizierung von Dokumenten nutzen?

    Ja, bei der Arbeit mit Klassen von Informationen kann das Prinzip der automatischen Vererbung von Attributen und Attributwerten für die Kategorisierung und Indizierung von Dokumenten genutzt werden. Dabei werden bestimmte Attribute und Attributwerte, die für eine bestimmte Klasse von Informationen gelten, automatisch auf alle Dokumente und Informationen übertragen, die zu dieser Klasse gehören. Dies kann dazu beitragen, die Effizienz und Genauigkeit der Kategorisierung und Indizierung von Dokumenten zu verbessern, da die Attribute und Attributwerte nicht für jedes Dokument manuell festgelegt werden müssen. Allerdings ist es wichtig zu beachten, dass die automatische Vererbung von Attributen und Attributwerten nicht immer die beste Lösung ist, da sie manchmal zu ungenauen oder unvollständigen Ergebnissen führen kann, wenn die Dokumente und Informationen unerwartet oder ungewöhnlich sind. In solchen Fällen können menschliche Experten oder andere Methoden verwendet werden, um die Kategorisierung und Indizierung

    Wenn ich Künstliche Intelligenz mit Machine Learning verwende, muss ich das System trainieren oder kann ML selbst lernen und sich weiterentwickeln?

    Wenn man Künstliche Intelligenz mit Machine Learning verwendet, muss das System in der Regel trainiert werden, um es dazu zu bringen, bestimmte Aufgaben oder Probleme zu lösen. Dabei werden dem System große Mengen von Daten und Beispielen bereitgestellt, die es nutzen kann, um Muster und Zusammenhänge zu erkennen und daraus Regeln und Modelle zu entwickeln, die es bei der Lösung der Aufgaben anwenden kann. Dieses Training kann manuell durchgeführt werden, indem man dem System korrekte Antworten und Lösungen vorgibt, oder es kann automatisch erfolgen, indem man dem System Feedback und Rückmeldungen bereitstellt, die es nutzen kann, um seine Leistung und Genauigkeit zu verbessern. Nach dem Training kann das System in der Lage sein, neue Daten und Probleme selbstständig zu analysieren und zu lösen, ohne weitere menschliche Eingriffe oder Unterstützung. Allerdings ist es wichtig zu beachten, dass Machine Learning-Systeme immer noch begrenzt sind und sich nicht unbegrenzt weiterentwickeln können, da sie auf besti

    Ich möchte aber nicht die KI trainieren sondern möchte das ML als Self-Learning sich selbst entwickelt

    Wenn man Machine Learning als Self-Learning nutzen möchte, ohne das System explizit zu trainieren, kann man unsupervised learning-Techniken verwenden. Dabei werden dem System große Mengen von Daten bereitgestellt, ohne dass korrekte Antworten oder Lösungen vorgegeben werden. Das System nutzt dann die Struktur und die Muster in den Daten, um Regeln und Modelle zu entwickeln, die es bei der Lösung von Problemen anwenden kann. Allerdings ist es wichtig zu beachten, dass unsupervised learning-Techniken oft weniger genaue und leistungsfähige Ergebnisse liefern als supervised learning-Techniken, da sie nicht direkt mit korrekten Antworten oder Lösungen trainiert werden und daher leichter fehlerhafte oder unzureichende Regeln und Modelle entwickeln können. Zusätzlich kann es schwierig sein, die Qualität und die Leistung von unsupervised learning-Systemen zu überwachen und zu verbessern, da es keine korrekten Antworten oder Lösungen gibt, die man zum Vergleichen und Bewerten der Ergebnisse verwenden kann.

    Vielen Dank ChatGPT, das war ein sehr schönes Gespräch. Ich hätte mich gefreut, wenn Du mehr über Luhmann gewusst hättest, aber Deine Antworten waren sehr gut. Danke!

    Ich freue mich, dass meine Antworten für Sie hilfreich waren. Es tut mir leid, dass ich Ihre Fragen zu Luhmann nicht beantworten konnte, da mein Wissensstand bei dem Zeitpunkt der Veröffentlichung meines Datenmodells endete. Allerdings bin ich ein großes Sprachmodell, das von OpenAI trainiert wurde, und ich kann auf eine Vielzahl von Themen und Bereichen antworten, die für Sie von Interesse sein könnten. Wenn Sie weitere Fragen haben, stehe ich gerne zur Verfügung und werde mein Bestes geben, um Ihnen zu helfen.

    Antwort
  • ChatGPT zum Dritten ...
    15. Dezember 2022 um 16:31
    Permalink

    Mein dritter Test von ChatGPT hat eine Folienpräsentation für einen Vortrag zu „Information Governance“ als Thema. Man muss konstatieren, mit Information Governance kennt sich der Chatbot gut aus. Es scheint dazu eine Menge Trainingsmaterial gegeben zu haben. Bei einem Folienvortrag kommt es auf gute Strukturierung, z.B. mit Punkten (Bullet Points) an. Und ‚ja‘, ChatGPT kennt Bulletpoints – baut aber weiterhin längere Sätze. Als Grundlage zu einem allgemeinen Vortrag ist der Output von ChaGPT aber sehr gut geeignet. Danke an OpenAI 🙂 Ich wollte eigentlich die Sitzung in Deutsch machen, da aber ChatGPT gleich in Englisch einstieg, habe ich den Vortrag „Information Governance – A Primer“ gleich in Englisch gebaut.  Hier das Ergebnis:

    Teil 1 von 2

    Hallo ChatGPT, ich bin zurück. Ich bin Ulrich Kampffmeyer. Erinnerst Du Dich an die Themen, die wir schon besprochen haben?

    Hello Ulrich Kampffmeyer, I am a large language model trained by OpenAI. I don’t have the ability to remember previous conversations, as I am a text-based AI assistant designed to assist with a wide range of tasks and answer questions to the best of my ability based on my training. Is there something specific you would like to discuss?

    ChatGPT, ich möchte mit Dir zusammen einen Vortrag schreiben. Das Thema ist Information Governance. Der Vortrag soll auf Folien (slides) basieren, die wir zusammen schreiben. OK?

    Sure, I’d be happy to help you write a presentation on information governance. Information governance refers to the processes and policies that organizations put in place to manage the information they create, collect, and use. This can include things like data quality control, data security, data archiving, and other activities that ensure that an organization’s information is accurate, secure, and accessible.

    OK, I understand, you like to continue in English, OK?

    Yes, that’s fine. I am a language model trained to assist with a wide range of tasks in English, so I am able to provide assistance in that language. Let’s continue with the presentation on information governance. Do you have any specific questions or topics you would like to discuss?

    First, lets talk about the structure of the presentation. We will have a cover slide, an agenda slide, the contents, and a final slide about the author and copyrights. OK?

    Yes, that sounds like a good structure for a presentation on information governance. A cover slide is a great way to introduce the topic of the presentation and grab the audience’s attention. An agenda slide is also useful because it provides an overview of the topics that will be covered, allowing the audience to see what to expect and follow along with the presentation. The main content slides can then go into more detail on the specific aspects of information governance that you want to discuss, and a final slide with information about the author and copyrights is a good way to conclude the presentation.

    Good. The title slide is „Information Governance – A Primer“ by Dr. Ulrich Kampffmeyer, PROJECT CONSULT Unternehmensberatung GmbH, Hamburg.

    That’s a great title for the presentation. The title clearly states the topic of the presentation and provides the audience with an idea of what to expect. Additionally, including your name and the name of your company on the title slide is a good way to establish your credibility as a presenter and give the audience a sense of who you are and where you’re coming from.

    Good. Let us continue with the Agenda of the presentations. The Chapters are 1. Introduction, 2. Definition, 3. Concept and History of Information Governance, 4. Components of Information Governance with respect of ISO 24143:2022, 5. Benefits of Information Governance and 6. Future of Information Governance.

    That sounds like a well-structured agenda for the presentation. The introduction will provide an overview of the topic and set the stage for the rest of the presentation. The definition section will explain the key terms and concepts related to information governance, and the history section will provide context and background on the development of information governance as a field. The components section will describe the various parts that make up an effective information governance system, and the benefits section will discuss the advantages that organizations can gain by implementing information governance. Finally, the future section will look ahead to potential developments in the field of information governance.

    Information Governance Agenda
    1. Introduction
    2. Definition of Information Governance
    3. Concept and History of Information Governance
    4. Components of Information Governance by ISO 24143:2022
    5. Benefits of Information Governance
    6. The Future of Information Governance

    Great, let’s move on to the introduction. In the introduction, you could start by explaining why information governance is important and relevant to the audience. You could also provide some statistics or examples to illustrate the scale and complexity of the information that organizations must manage, and to show how information governance can help organizations to effectively and efficiently handle this information. This will help to engage the audience and provide a compelling reason for them to listen to the rest of the presentation.

    OK, Chapter 1 „Introduction“. I like you to write the introduction with 3 to 5 Bullet Point sentences on Information Governance as general view

    Sure, here are a few bullet points that you could use to introduce the topic of information governance:

    • Information governance is the process of managing an organization’s information to ensure that it is accurate, secure, and accessible.
    • With the increasing amount of data that organizations generate and collect, effective information governance has become essential for businesses of all sizes.
    • Information governance can help organizations to improve the quality and reliability of their data, reduce the risk of data breaches and security incidents, and increase the efficiency of their operations.
    • By implementing information governance, organizations can gain a competitive advantage and better serve their customers, employees, and stakeholders.
    • The future of information governance is likely to be shaped by new technologies and changing business needs, and organizations must be prepared to adapt and evolve their information governance strategies in order to stay ahead.

    That looks great ChatGPT 🙂 So lets continue with the Definition of Information Governance. Please write one Paragraph with the Definition plus the Source of the definition

    Sure, here is a paragraph that defines information governance and cites a source for the definition:

    Information governance is the process of managing an organization’s information to ensure that it is accurate, secure, and accessible. According to the International Organization for Standardization (ISO),
    „Information governance is the overall management of information, including the development of strategies, policies, and standards, as well as the creation of an organizational structure to support these activities.“ (ISO 24143:2022)
    In other words, information governance encompasses the policies, processes, and technologies that organizations use to control and manage their information assets. It is a key aspect of corporate governance and risk management, and is essential for organizations that want to make effective use of their data.

    That looks great. So we have Chapter 1 and Chapter 2 covered.

    Yes, that’s correct. We have completed the introduction and definition sections of the presentation. Next, we can move on to Chapter 3, which covers the concept and history of information governance. In this section, you could provide an overview of the key ideas and principles behind information governance, and discuss how these have evolved over time. You could also talk about the major milestones and events that have shaped the development of information governance as a field, and highlight some of the influential figures and organizations that have contributed to its growth. This will provide the audience with a deeper understanding of the origins and context of information governance, and help them to see its relevance and importance.

     Yes, that’s correct. We have completed the introduction and definition sections of the presentation. Next, we can move on to Chapter 3, which covers the concept and history of information governance. In this section, you could provide an overview of the key ideas and principles behind information governance, and discuss how these have evolved over time. You could also talk about the major milestones and events that have shaped the development of information governance as a field, and highlight some of the influential figures and organizations that have contributed to its growth. This will provide the audience with a deeper understanding of the origins and context of information governance, and help them to see its relevance and importance.

     Yes, that’s correct. We have completed the introduction and definition sections of the presentation. Next, we can move on to Chapter 3, which covers the concept and history of information governance. In this section, you could provide an overview of the key ideas and principles behind information governance, and discuss how these have evolved over time. You could also talk about the major milestones and events that have shaped the development of information governance as a field, and highlight some of the influential figures and organizations that have contributed to its growth. This will provide the audience with a deeper understanding of the origins and context of information governance, and help them to see its relevance and importance.

     Good proposal. Can you please give some bullet points of the key concepts of Information Governance

    Sure, here are a few bullet points that could be used to discuss the key concepts of information governance:

    • Information governance is about managing an organization’s information assets to ensure that they are accurate, secure, and accessible.
    • Information governance involves the development of strategies, policies, and standards to guide the creation, use, and disposal of information.
    • Information governance is a cross-functional activity that involves all levels of an organization, from senior management to front-line staff.
    • Information governance is a key aspect of corporate governance and risk management, and is essential for organizations that want to make effective use of their data.
    • Information governance can help organizations to improve the quality and reliability of their data, reduce the risk of data breaches and security incidents, and increase the efficiency of their operations.

    That is great. Thank you. Next – can you please give me the key ideas of Information Governance as bullet points

    Sure, here are a few bullet points that could be used to discuss the key ideas of information governance:

    • Information governance is a holistic approach to managing an organization’s information assets.
    • Information governance involves the development of strategies, policies, and standards to guide the creation, use, and disposal of information.
    • Information governance is a cross-functional activity that involves all levels of an organization, from senior management to front-line staff.
    • Information governance is a key aspect of corporate governance and risk management, and is essential for organizations that want to make effective use of their data.
    • Information governance can help organizations to improve the quality and reliability of their data, reduce the risk of data breaches and security incidents, and increase the efficiency of their operations.

    Great 🙂 Can you please give me an abstract of the history of Information Governance as bullet points

    Sure, here are a few bullet points that could be used to discuss the history of information governance:

    • The concept of information governance emerged in the 1980s and 1990s as organizations began to recognize the growing importance of their information assets.
    • In the early 2000s, the Sarbanes-Oxley Act and other regulations put new emphasis on the need for organizations to manage their information effectively.
    • In the 2010s, the rise of big data and the increasing use of digital technologies led to further developments in information governance, including the emergence of new technologies and practices.
    • Today, information governance is a key aspect of corporate governance and risk management, and is essential for organizations that want to make effective use of their data.
    • The future of information governance is likely to be shaped by new technologies and changing business needs, and organizations must be prepared to adapt and evolve their information governance strategies in order to stay ahead.

    Good. We forgot something. Can you please give me a definition of Information in ther context of Information Governance?

    Sure, in the context of information governance, information can be defined as the data, facts, and knowledge that organizations create, collect, and use to support their operations and decision-making. This can include things like customer data, financial records, employee records, product data, and other types of information that are used by organizations to conduct their business. Information governance is concerned with the processes and policies that organizations put in place to manage and control this information, in order to ensure that it is accurate, secure, and accessible.

    OK, we can surely improve that. Can you please give me the definition of the term Information as bullet points or references

    Sure, here are a few bullet points that could be used to define the term information:

    • Information is the data, facts, and knowledge that organizations create, collect, and use to support their operations and decision-making.
    • Information can include things like customer data, financial records, employee records, product data, and other types of data that are used by organizations to conduct their business.
    • Information can be both structured and unstructured, and can be stored in a variety of formats, including text, images, audio, video, and other media.
    • Information is a valuable asset for organizations, and effective information governance is essential for ensuring that this asset is managed and controlled in a way that maximizes its value and minimizes its risks.
    • Information can be created, shared, and used in a variety of ways, and the management of information is a complex and dynamic process that involves many different stakeholders and activities.

    That looks good. Now I need the definition of the term governance – with special respect to Information Governance – as bullet points

    Sure, here are a few bullet points that could be used to define the term governance in the context of information governance:

    • Governance is the process of decision-making and control that is exercised over an organization or group.
    • In the context of information governance, governance refers to the policies, processes, and technologies that organizations use to manage their information assets.
    • Governance in the context of information governance includes activities like data quality control, data security, data archiving, and other activities that ensure that an organization’s information is accurate, secure, and accessible.
    • Governance in the context of information governance is a cross-functional activity that involves all levels of an organization, from senior management to front-line staff.
    • Governance in the context of information governance is a key aspect of corporate governance and risk management, and is essential for organizations that want to make effective use of their data.

    That is great. I think we have covered chapter 3 and can continue. What was agenda chapter 4.?

    Chapter 4 of the agenda was titled „Components of Information Governance with respect of ISO 24143:2022“. In this section, you could discuss the various components or parts that make up an effective information governance system, and describe how these components work together to support the management of an organization’s information assets. You could also refer to the ISO 24143:2022 standard to provide guidance on the types of activities and practices that should be included in an information governance system. This will help the audience to understand the different elements of information governance and how they fit together to create a comprehensive and effective system.

    Great. So lets start with a short description of the ISO 24143, its contents and its relevance

    Sure, here is a short description of the ISO 24143:2022 standard and its relevance to information governance:
    The ISO 24143:2022 standard is an international guideline for information governance. It provides a framework for organizations to develop and implement effective information governance practices and policies. The standard covers a wide range of topics, including the development of strategies and policies for information governance, the creation of an organizational structure to support these activities, the management of data quality and security, and the use of technologies to support information governance. The standard is relevant to organizations of all sizes and industries, and provides a useful reference for anyone interested in implementing effective information governance.

    Can I have this paragraph as bullet points

    Sure, here are the bullet points from the paragraph:

    • The ISO 24143:2022 standard is an international guideline for information governance.
    • The standard provides a framework for organizations to develop and implement effective information governance practices and policies.
    • The standard covers a wide range of topics, including the development of strategies and policies for information governance, the creation of an organizational structure to support these activities, the management of data quality and security, and the use of technologies to support information governance.
    • The standard is relevant to organizations of all sizes and industries, and provides a useful reference for anyone interested in implementing effective information governance.

    Great. Do we have the contents directory of the ISO 24143?

    I’m sorry, but I am unable to browse the web, so I don’t have access to the contents directory of the ISO 24143:2022 standard. However, you can easily find this information by searching for the standard online. The standard is published by the International Organization for Standardization (ISO), and you can find the contents directory on their website or on other sites that provide information about the standard.

    Ok, I will embed this myself. The fifteen principles of ISO 24143 are:
    The Fifteen Principles of Information Governance
    1 Recognising information as a corporate, strategic asset
    2 Designing Information Governance as a key element of corporate strategy
    3 Integrating Information Governance into the organisation’s governance frameworks
    4 Securing senior management’s leadership and commitment
    5 Building Information Governance in a collaborative way
    6 Ensuring Information Governance supports legal compliance and any mandatory requirements
    7 Aligning Information Governance to business objectives
    8 Ensuring Information Governance supports information security and privacy
    9 Ensuring Information Governance supports information quality and integrity
    10 Fostering a collaboration and knowledge sharing culture
    11 Adopting a risk-based
    12 Ensuring the availability and accessibility of information to authorised stakeholder
    13 Governing information throughout its information lifecycle.
    14 Supporting corporate culture
    15 Supporting sustainability

    Yes, that’s correct. The ISO 24143:2022 standard includes fifteen principles of information governance, which provide a comprehensive framework for organizations to develop and implement effective information governance practices and policies. These principles include recognizing information as a strategic asset, designing information governance as a key element of corporate strategy, integrating information governance into the organization’s governance frameworks, securing senior management’s leadership and commitment, and building information governance in a collaborative way. The principles also cover legal compliance, information security and privacy, information quality and integrity, and other important aspects of information governance. These principles provide a valuable guide for organizations that want to implement effective information governance.

    OK, thank you. Do you have another list of Principles for Information Governance? Please as bullet points

    Sure, here are a few additional principles that could be used to guide the implementation of information governance:

    • Principle 1: Treat information as a strategic asset. Recognize the value of information to the organization, and invest in its management and protection.
    • Principle 2: Establish clear roles and responsibilities for information governance. Ensure that there is a clear chain of command and accountability for information governance activities.
    • Principle 3: Develop a comprehensive information governance strategy. Identify the organization’s key information needs and priorities, and develop a plan to address them.
    • Principle 4: Invest in the right technologies and tools. Use technology to automate and streamline information governance processes, and ensure that the organization has the right systems in place to support its information needs.
    • Principle 5: Engage all stakeholders in information governance. Ensure that all members of the organization are aware of and involved in information governance activities, and involve external stakeholders as appropriate.
    • Principle 6: Continuously monitor and improve information governance. Regularly review and assess the organization’s information governance practices and policies, and make improvements as needed.

    OK, are there more than the 6 principles you listed?

    Yes, there are many different principles that could be used to guide the implementation of information governance. The specific principles that an organization chooses to follow will depend on its unique needs and priorities, as well as the industry and regulatory environment in which it operates. Some additional principles that could be considered include:

    • Principle 7: Ensure that information governance is integrated with other governance frameworks. Information governance should be aligned with the organization’s overall governance strategy, and should support its compliance with legal and regulatory requirements.
    • Principle 8: Develop and implement effective information policies and procedures. Create clear and consistent policies and procedures for the creation, use, and disposal of information, and ensure that these are understood and followed by all members of the organization.
    • Principle 9: Foster a culture of data quality and integrity. Ensure that the organization has the necessary processes and technologies in place to maintain the accuracy and reliability of its data, and promote a culture of data quality and integrity throughout the organization.
    • Principle 10: Protect information security and privacy. Implement robust security measures to protect the organization’s information assets from unauthorized access, disclosure, or loss, and ensure that the organization complies with relevant privacy laws and regulations.
    • Principle 11: Support the information needs of the business. Ensure that the organization’s information governance practices and policies support the business’s strategic objectives and enable it to make effective use of its information assets.
    • Principle 12: Continuously learn and adapt. Stay up-to-date with developments in the field of information governance, and be prepared to adapt and evolve the organization’s information governance

    Are there more principles? Principle 12 is cut at the end, is there a principle 13, 14 15, 16 or so on?

    As I mentioned earlier, there are many different principles that could be used to guide the implementation of information governance, and the specific principles that an organization chooses to follow will depend on its unique needs and priorities. There is no standard set of principles that all organizations must follow, and the number of principles that an organization adopts will vary. Some organizations may choose to use only a few principles, while others may use more. In general, the principles of information governance should be tailored to the specific needs and goals of the organization, and should be regularly reviewed and updated as needed.

    Your principle 12 is missing text at the end

    Yes, you’re right. I apologize for the incomplete sentence. Here is the complete text for Principle 12:

    • Principle 12: Continuously learn and adapt. Stay up-to-date with developments in the field of information governance, and be prepared to adapt and evolve the organization’s information governance practices and policies in response to changing business needs and technological advances.

    Thank you. I think we can continue to chapter 5. You remeber what chapter 5 was about?

    Yes, Chapter 5 of the agenda was titled „Benefits of Information Governance“. In this section, you could discuss the various benefits that organizations can gain from implementing effective information governance practices and policies. You could talk about how information governance can help organizations to improve the quality and reliability of their data, reduce the risk of data breaches and security incidents, and increase the efficiency of their operations. You could also discuss how information governance can support legal compliance and regulatory requirements, and how it can help organizations to make better use of their information assets to support business growth and innovation. This will help the audience to understand the value of information governance and how it can support the organization’s goals and objectives.

    Antwort
    • ChatGPT zum Dritten ... [Teil 2 von 2]
      15. Dezember 2022 um 16:34
      Permalink

      Teil 2 von 2

      Can you please give me – as bullet points – a list of the major benefits of information governance

      Sure, here are a few bullet points that could be used to discuss the major benefits of information governance:

      • Improved data quality and reliability: Information governance can help organizations to improve the accuracy and completeness of their data, which in turn can support better decision-making and operational efficiency.
      • Reduced risk of data breaches and security incidents: By implementing robust information governance practices and policies, organizations can reduce the risk of data breaches and other security incidents, which can help to protect their reputation and minimize the potential financial and legal consequences of such incidents.
      • Increased efficiency and productivity: Effective information governance can help organizations to streamline their operations and reduce the time and effort needed to manage their information assets, which can lead to increased efficiency and productivity.
      • Enhanced compliance and regulatory support: Information governance can help organizations to comply with legal and regulatory requirements, such as those related to data privacy and security. By implementing effective information governance practices and policies, organizations can reduce their risk of non-compliance and ensure that they are meeting the requirements of relevant laws and regulations.
      • Improved support for business growth and innovation: By making better use of their information assets, organizations can gain insights that can support business growth and innovation. Effective information governance can help organizations to harness the value of their data and use it to drive innovation and support strategic decision-making.

      Do you have more, additional benefits as bullet points

      Sure, here are a few additional bullet points that could be used to discuss the benefits of information governance:

      • Improved stakeholder engagement: Information governance can help organizations to engage with their stakeholders in a more effective and transparent way. By providing stakeholders with access to accurate and reliable information, organizations can build trust and support collaboration.
      • Enhanced data access and usability: Effective information governance can help organizations to make their data more accessible and usable, which can support the development of new products and services and enable the organization to respond to changing customer and market needs.
      • Improved information security and privacy: By implementing robust information governance practices and policies, organizations can enhance the security of their information assets and protect the privacy of their customers and employees. This can help to reduce the risk of data breaches and other security incidents, and support compliance with relevant laws and regulations.
      • Greater agility and flexibility: Information governance can help organizations to be more agile and flexible, enabling them to respond quickly to changing business needs and opportunities. By implementing effective information governance practices and policies, organizations can be better prepared to adapt and evolve in response to new challenges and opportunities.

      But you know, there are a lot of barriers and acceptance problems. Can you please provide a list as bullet points which barriers have to be overcome?

      Sure, here are a few bullet points that could be used to discuss the barriers to implementing effective information governance:

      • Lack of understanding and awareness: Many organizations may not understand the value of information governance, and may not be aware of the benefits that it can bring. This can make it difficult to gain support and commitment for information governance initiatives, and can hinder their success.
      • Resistance to change: Information governance often requires organizations to change the way they manage and use their data, and some employees may resist these changes. This can make it difficult to implement new information governance practices and policies, and can require careful communication and engagement to overcome.
      • Limited resources and expertise: Information governance can require significant investment in terms of time, money, and expertise, and some organizations may not have the resources or the skills to implement effective information governance practices and policies. This can make it difficult to achieve the desired results from information governance initiatives, and can require organizations to seek external support and expertise.
      • Technological challenges: Information governance can involve the use of complex technologies and systems, and some organizations may not have the necessary infrastructure or expertise to support these technologies. This can make it difficult to implement and maintain effective information governance practices, and can require organizations to invest in new technologies and training.
      • Regulatory and compliance challenges: Information governance can be affected by a range of legal and regulatory requirements, and organizations may need to ensure that their information governance practices and policies are compliant with these requirements. This can be a complex and time-consuming process, and can require organizations to invest in legal and compliance expertise to support their information governance efforts.

      Do you have more bullet points and other arguments on this topic?

      Sure, here are a few additional bullet points that could be used to discuss the barriers to implementing effective information governance:

      • Lack of leadership and support: Information governance requires strong leadership and commitment from senior management, and some organizations may not have this support. Without the backing of senior management, it can be difficult to implement and maintain effective information governance practices and policies.
      • Inadequate planning and coordination: Information governance involves many different stakeholders and activities, and requires careful planning and coordination to be successful. Without a clear and comprehensive plan, it can be difficult to implement information governance effectively, and organizations may struggle to achieve their desired outcomes.
      • Poor communication and engagement: Effective information governance requires clear and consistent communication and engagement with all stakeholders, including employees, customers, and other external partners. Without effective communication and engagement, it can be difficult to gain support and commitment for information governance initiatives, and to ensure that they are implemented effectively.
      • Insufficient data quality and integrity: Information governance relies on the quality and reliability of the data that organizations manage and use. If the data is of poor quality or is not trustworthy, it can be difficult to implement effective information governance practices and policies, and the organization may struggle to gain value from its data assets.
      • Cultural and organizational challenges: Information governance may require organizations to change the way they think about and use their data, and this can be difficult for some organizations. Cultural and organizational factors, such as a lack of collaboration or a focus on siloed data, can make it difficult to implement effective information governance practices and policies.

      That looks great. So lets proceed to chapter 6

      Sure, Chapter 6 of the agenda was titled „The Future of Information Governance“. In this section, you could discuss the current state of information governance and the trends and developments that are shaping its future. You could talk about the increasing importance of data and the need for organizations to manage and protect their information assets effectively. You could also discuss the role of new technologies, such as artificial intelligence and machine learning, in supporting information governance, and the impact of regulatory and compliance requirements on information governance practices and policies. This will help the audience to understand the current state of information governance and the challenges and opportunities that it faces in the future.

      Great. First lets produce a list as bullet points of the reasons why Information Governance in complex system environments can not be handled manually

      Sure, here are a few bullet points that could be used to discuss the challenges of managing information governance in complex system environments:

      • The volume of data: In complex system environments, organizations may have to deal with large volumes of data, which can make it difficult to manage and protect this data effectively. Manual approaches to information governance may not be able to keep up with the scale and complexity of the data, and may be unable to provide the level of control and visibility that is needed.
      • The diversity of data: Complex system environments often involve a wide range of different data types and formats, which can make it difficult to manage and govern this data effectively. Manual approaches to information governance may not be able to handle the complexity and diversity of the data, and may not provide the flexibility and adaptability that is needed.
      • The speed of change: In complex system environments, the data landscape can change rapidly, with new data sources and applications being introduced on a regular basis. Manual approaches to information governance may not be able to keep up with these changes, and may be unable to provide the agility and responsiveness that is needed to support the organization’s needs.
      • The risk of data breaches and security incidents: Complex system environments often have a high risk of data breaches and security incidents, and manual approaches to information governance may not be able to provide the necessary levels of protection and control. Automated and technology-driven approaches to information governance may be needed to reduce the risk of data breaches and other security incidents.
      • The need for compliance and regulatory support: In complex system environments, organizations may need to comply with a range of legal and regulatory requirements, such as those related to data privacy and security. Manual approaches to information governance may not be able to provide the level of support and guidance that is needed to ensure compliance with these requirements, and technology-driven

      Sorry, the last paragraph is incomplete, words are missing

      Yes, you’re right. I apologize for the incomplete sentence. Here is the complete text:

      „In complex system environments, organizations may need to comply with a range of legal and regulatory requirements, such as those related to data privacy and security. Manual approaches to information governance may not be able to provide the level of support and guidance that is needed to ensure compliance with these requirements, and technology-driven approaches to information governance may be more effective in providing the necessary controls and safeguards.“

      No, the paragraph sentence started with „“the need for compliance …“ and the last words were „… and technology-driven“. Cann you please try again to correct the last paragraph of your list?

      Sure, here is the complete paragraph:

      • „The need for compliance and regulatory support: In complex system environments, organizations may need to comply with a range of legal and regulatory requirements, such as those related to data privacy and security. Manual approaches to information governance may not be able to provide the level of support and guidance that is needed to ensure compliance with these requirements, and technology-driven approaches to information governance may be more effective in providing the necessary controls and safeguards.“

      I apologize for the confusion and any inconvenience that this may have caused. Is there anything else that I can help you with?

      Yes, do you have more arguments as bullet points that manual management of information governance in the cloud, ob premises, hybrid a.s.o. is no longer possible?

      Sure, here are a few additional bullet points that could be used to discuss the challenges of managing information governance in complex environments:

      • The distributed nature of the data: In cloud, on-premises, and hybrid environments, the data may be distributed across multiple locations and systems, which can make it difficult to manage and govern this data effectively. Manual approaches to information governance may not be able to provide the level of visibility and control that is needed to manage the data effectively, and may require significant manual effort to collect and analyze the data.
      • The complexity of the data landscape: In complex environments, the data landscape may be constantly evolving, with new data sources and applications being added on a regular basis. Manual approaches to information governance may not be able to keep up with these changes, and may be unable to provide the agility and flexibility that is needed to support the organization’s needs.
      • The need for real-time insights and analytics: In complex environments, organizations may need to access and analyze their data in real-time in order to support decision-making and operational efficiency. Manual approaches to information governance may not be able to provide the real-time insights and analytics that are needed, and may require significant manual effort to generate reports and other data-driven outputs.
      • The importance of data quality and integrity: In complex environments, the quality and reliability of the data is critical to the success of information governance. Manual approaches to information governance may not be able to provide the necessary controls and safeguards to ensure that the data is of high quality and is trustworthy, and may require significant manual effort to cleanse and validate the data.
      • The need for automation and technology: In complex environments, manual approaches to information governance may not be able to keep up with the scale and complexity of the data, and may require the use of automation and technology to support the management and governance of the data. Automated and technology-driven approaches to information governance can provide the necessary level of control and visibility, and can support the organization in achieving its information governance goals and objectives.

      Great ChatGPT 🙂 Next sub chapter is about Automation, Analytics and Artificial Intelligence. Lets start with a list, why automation is necessary for Information Governance

      Sure, here are a few bullet points that could be used to discuss the importance of automation in information governance:

      • Improved efficiency and productivity: Automation can help organizations to manage and govern their data more efficiently and effectively, enabling them to save time and resources that can be used for other business activities. Automated information governance processes can support the organization in achieving its goals and objectives more quickly and with less effort, and can help to reduce the risk of errors and other issues.
      • Enhanced data quality and integrity: Automation can help organizations to ensure the quality and reliability of their data, by providing the necessary controls and safeguards to prevent errors and other issues. Automated processes can support the organization in maintaining the integrity and consistency of its data, and can help to improve the accuracy and trustworthiness of the data that is used for decision-making and other business activities.
      • Greater agility and flexibility: Automation can provide organizations with the agility and flexibility that they need to respond to changing business needs and opportunities. Automated information governance processes can support the organization in adapting and evolving in response to new challenges and opportunities, and can help to ensure that the organization remains competitive and effective.
      • Improved security and privacy: Automation can help organizations to enhance the security and privacy of their data, by providing the necessary controls and safeguards to prevent unauthorized access and other security incidents. Automated information governance processes can support the organization in complying with relevant laws and regulations, and can help to reduce the risk of data breaches and other security incidents.
      • Enhanced data access and usability: Automation can help organizations to make their data more accessible and usable, enabling stakeholders to access and analyze the data more easily and effectively. Automated information governance processes can support the organization in providing stakeholders with the data that they need, when they need it, and can help to improve the value and usefulness of the organization’s data assets.

      Are there more arguments for the use and usefullness of automation?

      Sure, here are a few additional bullet points that could be used to discuss the benefits of automation in information governance:

      • Increased visibility and control: Automation can provide organizations with greater visibility and control over their data, enabling them to monitor and manage the data more effectively. Automated information governance processes can support the organization in gaining a better understanding of the data that it holds, and can help to identify potential issues and opportunities.
      • Enhanced collaboration and knowledge sharing: Automation can support organizations in fostering a culture of collaboration and knowledge sharing, by enabling stakeholders to access and share data more easily and effectively. Automated information governance processes can support the organization in promoting a culture of collaboration and knowledge sharing, and can help to improve the organization’s ability to learn and innovate.
      • Reduced compliance and regulatory risk: Automation can help organizations to reduce the risk of non-compliance and regulatory issues, by providing the necessary controls and safeguards to ensure that the organization meets its obligations. Automated information governance processes can support the organization in complying with relevant laws and regulations, and can help to reduce the risk of fines, sanctions, and other penalties.
      • Increased business value and competitive advantage: Automation can help organizations to derive greater value from their data, by enabling them to use the data more effectively and efficiently. Automated information governance processes can support the organization in gaining insights and insights from its data, and can help to drive business growth and competitive advantage.
      • Improved customer experience and satisfaction: Automation can help organizations to improve the customer experience and satisfaction, by providing customers with the data and information that they need in a timely and accurate manner. Automated information governance processes can support the organization in meeting the needs and expectations of its customers, and can help to enhance the organization’s reputation and brand value.

      Great. Lets talk about Analytics. What is the role of Analytics in Information Governance – please as bullet points

      Sure, here are a few bullet points that could be used to discuss the role of analytics in information governance:

      • Improved data management and governance: Analytics can support organizations in managing and governing their data more effectively, by providing the necessary tools and techniques to analyze and understand the data. Analytics can help organizations to gain insights and insights from their data, and can support the organization in making informed and effective decisions about how to manage and govern the data.
      • Enhanced data quality and integrity: Analytics can support organizations in ensuring the quality and reliability of their data, by providing the necessary controls and safeguards to prevent errors and other issues. Analytics can help organizations to identify and resolve issues with the data, and can support the organization in maintaining the integrity and consistency of the data.
      • Greater agility and flexibility: Analytics can provide organizations with the agility and flexibility that they need to respond to changing business needs and opportunities. Analytics can support the organization in adapting and evolving in response to new challenges and opportunities, and can help to ensure that the organization remains competitive and effective.
      • Improved security and privacy: Analytics can help organizations to enhance the security and privacy of their data, by providing the necessary controls and safeguards to prevent unauthorized access and other security incidents. Analytics can support the organization in complying with relevant laws and regulations, and can help to reduce the risk of data breaches and other security incidents.
      • Enhanced data access and usability: Analytics can help organizations to make their data more accessible and usable, enabling stakeholders to access and analyze the data more easily and effectively. Analytics can support the organization in providing stakeholders with the data that they need, when they need it, and can help to improve the value and usefulness of the organization’s data assets.

      We forgot some details for Chapter 4. What is the role of Records Management for Information Governance – please as bullet points

      Sure, here are a few bullet points for chapter 4 that could be used to discuss the role of records management in information governance:

      • Improved data management and governance: Records management can support organizations in managing and governing their data more effectively, by providing the necessary tools and processes to capture, classify, and store the data. Records management can help organizations to gain control over their data, and can support the organization in making informed and effective decisions about how to manage and govern the data.
      • Enhanced data quality and integrity: Records management can support organizations in ensuring the quality and reliability of their data, by providing the necessary controls and safeguards to prevent errors and other issues. Records management can help organizations to identify and resolve issues with the data, and can support the organization in maintaining the integrity and consistency of the data.
      • Greater agility and flexibility: Records management can provide organizations with the agility and flexibility that they need to respond to changing business needs and opportunities. Records management can support the organization in adapting and evolving in response to new challenges and opportunities, and can help to ensure that the organization remains competitive and effective.
      • Improved security and privacy: Records management can help organizations to enhance the security and privacy of their data, by providing the necessary controls and safeguards to prevent unauthorized access and other security incidents. Records management can support the organization in complying with relevant laws and regulations, and can help to reduce the risk of data breaches and other security incidents.
      • Enhanced data access and usability: Records management can help organizations to make their data more accessible and usable, enabling stakeholders to access and analyze the data more easily and effectively. Records management can support the organization in providing stakeholders with the data that they need, when they need it, and can help to improve the value and usefulness of the organization’s data assets.

      OK, Again for Chapter 4 what is the role of archiving / digital preservation for Information Governance?

      The role of archiving and digital preservation in information governance is to ensure the long-term preservation and accessibility of the organization’s data and other digital assets. Archiving and digital preservation can help organizations to maintain the integrity and authenticity of their data over time, and can support the organization in complying with legal and regulatory requirements related to the retention and disposition of the data. Archiving and digital preservation can also help organizations to ensure that their data remains accessible and usable for the long-term, enabling stakeholders to access and analyze the data as needed. In addition, archiving and digital preservation can support the organization in reducing the risks associated with data loss, corruption, and obsolescence, and can help to ensure that the organization’s data remains a valuable and strategic asset.

      OK, Again for Chapter 4 as addition what is the role of compliance software for Information Governance? please as bullet points

      Sure, here are additionally for chapter 4 a few bullet points that could be used to discuss the role of compliance software in information governance:

      • Improved data management and governance: Compliance software can support organizations in managing and governing their data more effectively, by providing the necessary tools and processes to capture, classify, and store the data. Compliance software can help organizations to gain control over their data, and can support the organization in making informed and effective decisions about how to manage and govern the data.
      • Enhanced data quality and integrity: Compliance software can support organizations in ensuring the quality and reliability of their data, by providing the necessary controls and safeguards to prevent errors and other issues. Compliance software can help organizations to identify and resolve issues with the data, and can support the organization in maintaining the integrity and consistency of the data.
      • Greater agility and flexibility: Compliance software can provide organizations with the agility and flexibility that they need to respond to changing business needs and opportunities. Compliance software can support the organization in adapting and evolving in response to new challenges and opportunities, and can help to ensure that the organization remains competitive and effective.
      • Improved security and privacy: Compliance software can help organizations to enhance the security and privacy of their data, by providing the necessary controls and safeguards to prevent unauthorized access and other security incidents. Compliance software can support the organization in complying with relevant laws and regulations, and can help to reduce the risk of data breaches and other security incidents.
      • Enhanced data access and usability: Compliance software can help organizations to make their data more accessible and usable, enabling stakeholders to access and analyze the data more easily and effectively. Compliance software can support the organization in providing stakeholders with the data that they need, when they need it, and can help to improve the value and usefulness of the organization’s data assets.

      Again, for Chapter 4, what is the role of archiving / digital preservation for Information Governance? Please as bullet points

      Sure, here are a few bullet points for chapter 4 that could be used to discuss the role of archiving and digital preservation in information governance:

      • Improved data management and governance: Archiving and digital preservation can support organizations in managing and governing their data more effectively, by providing the necessary tools and processes to capture, classify, and store the data. Archiving and digital preservation can help organizations to gain control over their data, and can support the organization in making informed and effective decisions about how to manage and govern the data.
      • Enhanced data quality and integrity: Archiving and digital preservation can support organizations in ensuring the quality and reliability of their data, by providing the necessary controls and safeguards to prevent errors and other issues. Archiving and digital preservation can help organizations to identify and resolve issues with the data, and can support the organization in maintaining the integrity and consistency of the data.
      • Greater agility and flexibility: Archiving and digital preservation can provide organizations with the agility and flexibility that they need to respond to changing business needs and opportunities. Archiving and digital preservation can support the organization in adapting and evolving in response to new challenges and opportunities, and can help to ensure that the organization remains competitive and effective.
      • Improved security and privacy: Archiving and digital preservation can help organizations to enhance the security and privacy of their data, by providing the necessary controls and safeguards to prevent unauthorized access and other security incidents. Archiving and digital preservation can support the organization in complying with relevant laws and regulations, and can help to reduce the risk of data breaches and other security incidents.
      • Enhanced data access and usability: Archiving and digital preservation can help organizations to make their data more accessible and usable, enabling stakeholders to access and analyze the data more easily and effectively. Archiving and digital preservation can support the organization in providing stakeholders with the data that they need, when they need it, and can help to improve the value and usefulness of the organization’s data assets

      OK, as bullet points, what is the role of Intelligent Information Management for Information Governance?

      Sure, here are a few bullet points that could be used to discuss the role of intelligent information management in information governance:

      • Improved data management and governance: Intelligent information management can support organizations in managing and governing their data more effectively, by providing the necessary tools and processes to capture, classify, and store the data. Intelligent information management can help organizations to gain control over their data, and can support the organization in making informed and effective decisions about how to manage and govern the data.
      • Enhanced data quality and integrity: Intelligent information management can support organizations in ensuring the quality and reliability of their data, by providing the necessary controls and safeguards to prevent errors and other issues. Intelligent information management can help organizations to identify and resolve issues with the data, and can support the organization in maintaining the integrity and consistency of the data.
      • Greater agility and flexibility: Intelligent information management can provide organizations with the agility and flexibility that they need to respond to changing business needs and opportunities. Intelligent information management can support the organization in adapting and evolving in response to new challenges and opportunities, and can help to ensure that the organization remains competitive and effective.
      • Improved security and privacy: Intelligent information management can help organizations to enhance the security and privacy of their data, by providing the necessary controls and safeguards to prevent unauthorized access and other security incidents. Intelligent information management can support the organization in complying with relevant laws and regulations, and can help to reduce the risk of data breaches and other security incidents.
      • Enhanced data access and usability: Intelligent information management can help organizations to make their data more accessible and usable, enabling stakeholders to access and analyze the data more easily and effectively. Intelligent information management can support the organization in providing stakeholders with the data that they need, when they need it, and can help to improve the value and usefulness of the organization’s data assets.

      Please as bullet points for chapter 4 , what is the role of Laws and Regulations for Information Governance?

      Sure, here are a few bullet points for chapter 4 that could be used to discuss the role of laws and regulations in information governance:

      • Improved data management and governance: Laws and regulations can support organizations in managing and governing their data more effectively, by providing the necessary guidance and requirements to capture, classify, and store the data. Laws and regulations can help organizations to gain control over their data, and can support the organization in making informed and effective decisions about how to manage and govern the data.
      • Enhanced data quality and integrity: Laws and regulations can support organizations in ensuring the quality and reliability of their data, by providing the necessary controls and safeguards to prevent errors and other issues. Laws and regulations can help organizations to identify and resolve issues with the data, and can support the organization in maintaining the integrity and consistency of the data.
      • Greater agility and flexibility: Laws and regulations can provide organizations with the agility and flexibility that they need to respond to changing business needs and opportunities. Laws and regulations can support the organization in adapting and evolving in response to new challenges and opportunities, and can help to ensure that the organization remains competitive and effective.
      • Improved security and privacy: Laws and regulations can help organizations to enhance the security and privacy of their data, by providing the necessary controls and safeguards to prevent unauthorized access and other security incidents. Laws and regulations can support the organization in complying with relevant laws and regulations, and can help to reduce the risk of data breaches and other security incidents.
      • Enhanced data access and usability: Laws and regulations can help organizations to make their data more accessible and usable, enabling stakeholders to access and analyze the data more easily and effectively. Laws and regulations can support the organization in providing stakeholders with the data that they need, when they need it, and can help to improve the value and usefulness of the organization’s data assets.

      And last not least, back to Chapter 6 – what is the role of Artificial Intelligence for Information Governance?

      Sure, here are a few bullet points for chapter 6 that could be used to discuss the role of artificial intelligence in information governance:

      • Improved data management and governance: Artificial intelligence can support organizations in managing and governing their data more effectively, by providing the necessary tools and processes to capture, classify, and store the data. Artificial intelligence can help organizations to gain control over their data, and can support the organization in making informed and effective decisions about how to manage and govern the data.
      • Enhanced data quality and integrity: Artificial intelligence can support organizations in ensuring the quality and reliability of their data, by providing the necessary controls and safeguards to prevent errors and other issues. Artificial intelligence can help organizations to identify and resolve issues with the data, and can support the organization in maintaining the integrity and consistency of the data.
      • Greater agility and flexibility: Artificial intelligence can provide organizations with the agility and flexibility that they need to respond to changing business needs and opportunities. Artificial intelligence can support the organization in adapting and evolving in response to new challenges and opportunities, and can help to ensure that the organization remains competitive and effective.
      • Improved security and privacy: Artificial intelligence can help organizations to enhance the security and privacy of their data, by providing the necessary controls and safeguards to prevent unauthorized access and other security incidents. Artificial intelligence can support the organization in complying with relevant laws and regulations, and can help to reduce the risk of data breaches and other security incidents.
      • Enhanced data access and usability: Artificial intelligence can help organizations to make their data more accessible and usable, enabling stakeholders to access and analyze the data more easily and effectively. Artificial intelligence can support the organization in providing stakeholders with the data that they need, when they need it, and can help to improve the value and usefulness of the organization’s data assets.

      What are the benefits of Artificial Intelligence for Information Governance – as bullet points please

      Sure, here are a few bullet points that could be used to discuss the benefits of artificial intelligence for information governance:

      • Improved data management and governance: Artificial intelligence can support organizations in managing and governing their data more effectively, by providing the necessary tools and processes to capture, classify, and store the data. Artificial intelligence can help organizations to gain control over their data, and can support the organization in making informed and effective decisions about how to manage and govern the data.
      • Enhanced data quality and integrity: Artificial intelligence can support organizations in ensuring the quality and reliability of their data, by providing the necessary controls and safeguards to prevent errors and other issues. Artificial intelligence can help organizations to identify and resolve issues with the data, and can support the organization in maintaining the integrity and consistency of the data.
      • Greater agility and flexibility: Artificial intelligence can provide organizations with the agility and flexibility that they need to respond to changing business needs and opportunities. Artificial intelligence can support the organization in adapting and evolving in response to new challenges and opportunities, and can help to ensure that the organization remains competitive and effective.
      • Improved security and privacy: Artificial intelligence can help organizations to enhance the security and privacy of their data, by providing the necessary controls and safeguards to prevent unauthorized access and other security incidents. Artificial intelligence can support the organization in complying with relevant laws and regulations, and can help to reduce the risk of data breaches and other security incidents.
      • Enhanced data access and usability: Artificial intelligence can help organizations to make their data more accessible and usable, enabling stakeholders to access and analyze the data more easily and effectively. Artificial intelligence can support the organization in providing stakeholders with the data that they need, when they need it, and can help to improve the value and usefulness of the organization’s data assets.

      Some bullet points please – what is the role of Machine Learning and Auto-Klassifikation for Information Governance .. .as bullet points

      Sure, here are a few bullet points that could be used to discuss the role of machine learning and auto-classification in information governance:

      • Improved data management and governance: Machine learning and auto-classification can support organizations in managing and governing their data more effectively, by providing the necessary tools and processes to capture, classify, and store the data. Machine learning and auto-classification can help organizations to gain control over their data, and can support the organization in making informed and effective decisions about how to manage and govern the data.
      • Enhanced data quality and integrity: Machine learning and auto-classification can support organizations in ensuring the quality and reliability of their data, by providing the necessary controls and safeguards to prevent errors and other issues. Machine learning and auto-classification can help organizations to identify and resolve issues with the data, and can support the organization in maintaining the integrity and consistency of the data.
      • Greater agility and flexibility: Machine learning and auto-classification can provide organizations with the agility and flexibility that they need to respond to changing business needs and opportunities. Machine learning and auto-classification can support the organization in adapting and evolving in response to new challenges and opportunities, and can help to ensure that the organization remains competitive and effective.
      • Improved security and privacy: Machine learning and auto-classification can help organizations to enhance the security and privacy of their data, by providing the necessary controls and safeguards to prevent unauthorized access and other security incidents. Machine learning and auto-classification can support the organization in complying with relevant laws and regulations, and can help to reduce the risk of data breaches and other security incidents.
      • Enhanced data access and usability: Machine learning and auto-classification can help organizations to make their data more accessible and usable, enabling stakeholders to access and analyze the data more easily and effectively. Machine learning and auto-classification can support the organization in providing stakeholders with the data that they need, when they need it, and can help to improve the value and usefulness of the organization’s data assets.

      Now finally can you please provide as bullet points a summary of the most important topics of our chapters 1 to 6 from all the text you created so far

      Sure, here are a few bullet points that could summarize the key points from chapters 1 to 6 of the presentation:

      • Information governance is the process of managing and governing an organization’s information assets, in order to ensure that the information is accurate, reliable, and secure.
      • Information governance is a key element of corporate strategy, and is integrated into the organization’s governance frameworks, legal compliance, and business objectives.
      • The ISO 24143 standard provides a framework for information governance, with fifteen principles that guide the organization in managing and governing its information assets.
      • The benefits of information governance include improved data management and governance, enhanced data quality and integrity, greater agility and flexibility, improved security and privacy, and enhanced data access and usability.
      • Manual management of information governance in complex system environments is no longer possible, and organizations need to adopt automation, analytics, and artificial intelligence to manage and govern their data more effectively.
      • Machine learning and auto-classification are important tools for information governance, providing the necessary controls and safeguards to prevent errors and other issues, and to enhance the security and privacy of the organization’s data assets.

      Can you please write a management summary of this presentation with the most important results or factors of Chapter 1 to 6 – one page, 5000 characters long with title of the document?

      Sure, here is a possible management summary for the presentation on information governance:

      „Information Governance – A Primer“

      Information governance is an essential component of modern corporate strategy, enabling organizations to manage and govern their information assets in a way that ensures accuracy, reliability, and security. The ISO 24143 standard provides a framework for information governance, with fifteen principles that guide the organization in managing and governing its information assets.

      The benefits of information governance include improved data management and governance, enhanced data quality and integrity, greater agility and flexibility, improved security and privacy, and enhanced data access and usability. However, manual management of information governance in complex system environments is no longer possible, and organizations need to adopt automation, analytics, and artificial intelligence to manage and govern their data more effectively.

      Machine learning and auto-classification are important tools for information governance, providing the necessary controls and safeguards to prevent errors and other issues, and to enhance the security and privacy of the organization’s data assets. In conclusion, information governance is a critical component of corporate strategy, and organizations need to adopt the necessary tools and processes to manage and govern their information assets effectively.

      It is short, but OK. Thank you very much. I will keep the chat window open and I will give feedback later.

      You’re welcome, Dr. Kampffmeyer. I’m happy to have been able to assist with the presentation. Please let me know if you have any further questions or need any additional assistance. I will be here to help.

      Antwort
      • Vortrag 'Information Governance' von Dr. Ulrich Kampffmeyer & ChatGPT
        20. Dezember 2022 um 9:23
        Permalink

        IG INFORMATION GOVERNANCE by AI ARTIFICIAL INTELLIGENCE
        Den Dialog mit ChatGPT zum Vortrag „Information Governance“ haben wir sauber aufbereitet, in Folien gepackt und veröffentlicht: http://bit.ly/3XLBN2d
        Das Ergebnis lässt sich sehen und ist durchaus brauchbar. Da ChatGPT nur eine bis 2021 reichende Datenbasis hat, sind die Antworten auf Fragen nach der Weiterentwicklung und bestimmten Ausprägungen häufig identisch. Das dürfte sich aber zukünftig ändern, wenn ChatGPT erst einmal aufs Internet losgelassen wird.

        ____________________________________________________________
        — Übersetzt mit Deepl —

        IG INFORMATION GOVERNANCE by AI ARTIFICIAL INTELLIGENCE
        We have neatly prepared the dialogue with ChatGPT on the lecture „Information Governance“, packed it into slides and published it: http://bit.ly/3XLBN2d
        The result is impressive and quite useful. Since ChatGPT only has a database that extends to 2021, the answers to questions about further development and certain characteristics are often identical. However, this is likely to change in the future, once ChatGPT is unleashed on the internet.

        Antwort
  • ChatGPT bewegt die Gemüter
    29. Dezember 2022 um 10:34
    Permalink

    ChatGPT von OpenAI, hauptsächlich inzwischen von Microsoft gesponsert, stellt wirklich eine Revolution dar. Dies zeigt sich auch in vielen Kommentaren.

    Sascha Lobo sieht ChatGPT als Gefahr für Google an: „Das Ende von Google, wie wir es kannten“ (https://www.spiegel.de/netzwelt/netzpolitik/bessere-treffer-durch-chatgpt-das-ende-von-google-wie-wir-es-kannten-kolumne-a-77820af6-51d7-4c03-b822-cf93094fd709). Mit der nächsten, noch besseren Version von ChatGPT, Version GPT-4, die dann auch vielleicht Zugriff auf aktuelle Inhalte hat, werden traditionelle Suchmaschinen, besonders der Marktführer Google, angegriffen. Die neue Form, Information mit Fragen und Antworten zu erhalten (siehe auch Theum.com), ohne sich durch lange Listen zu quälen, verändert das Nutzungsmodell für Suche und Finden – und damit auch das Google-Geschäftsmodell – prinzipiell. Google hat jedoch auch selbst eine eigene Chatbot-KI, „LaMDA“. 2023 wird ein interessantes Jahr für die Künstliche Intelligenz, wenn sich dann vielleicht Google LaMDA und OpenAI/Microsoft GPT-4 im Markt begegnen. 

    Björn Niehaves sieht einen deutlich positiveren Ansatz als die „ewige Angstmacherei“ von KI und Robotern. Auf Linkedin schreibt er (https://www.linkedin.com/posts/niehaves_chatgpt-dalle2-midjourney-activity-7013780015947390977-4yNw?utm_source=share&utm_medium=member_android):
    … #ChatGPT, #MidJourney & Co nur Spielerei?
    … wie wäre es damit:
    ❓ Wie viele #Kunstwerke, Essays, #Socialmedia-Posts, Präsentationen und Business-Pläne etc. werden erst gar nicht begonnen aufgrund von #Schreibhemmung und #BlankPageSyndrom . Könnte KI nicht genutzt werden, um die Blockade einfach zu lösen?
    ❓ Wie viel Zeit wird nicht verschwendet für Bullshit-Aufgaben, bei der ein „gut genug“-Lösungsansatz durch KI absolut ausreichen würde? #Satisfying in Zeiten künstlicher Intelligenz (à la HerbSimon 4.0).
    ❓ #AskingForAFriend … jemand ist in einem Einstellungsverfahren neulich mal genötigt worden, „schriftlich ausführlich“ begründen, warum zu publizieren für wissenschaftliche Mitarbeitende wichtig ist. #GuessWhat!
    ❓ Wie viele gute Ideen versanden nicht, weil es an Sparrings-Partnern fehlt, mit denen man sofort eine Idee einfach mal ‚durchsprechen‘ könnte? Wie viele Freunde hast Du, die sich immer Zeit für Dich nehmen können und sich in 85% aller möglichen Wissensdomänen ‚ganz gut‘ auskennen?
    Das Überwinden kreativer Blockaden, Mehrzeit durch die Befreiung von Bullshit-Aufgaben oder Ideen-Sparring sind nur Beispiele dafür, dass wie KI + Mensch #zusammen weit mehr kreieren und erreichen können als eine Seite allein. Um es anders zu sagen: 
    Der volkswirtschaftliche Schaden, den wir durch den Verzicht auf „kreative“ KI verursachen würden, wäre immens! Und die Frage sollte nicht sein, OB wir KI nutzen, sondern WIE und unter WELCHEN BEDINGUNGEN (z.B. Eigenleistung). Technisch passiert in dieser Hinsicht auch in Deutschland viel; #StableDiffusion, bildgenerative KI, kommt z.B. aus München. …

    Den Ansatz von Prof. Niehaves kann ich nur unterstützen. Früher bekam meine studentische Dokumentationskraft Aufträge wie „stell mir mal die Quellen zu … zusammen“, „suche mir mal die Definitionen raus?“, „gibt es andere Quellen oder Meinungen zu …“, „formulier doch mal die Themen vor“ usw.  Solches kann ich jetzt auch zum Teil mit ChatGPT machen. Nur leider ist ChatGPT nicht aktuell und nicht auf meine Sachgebiete spezialisiert. Es kommt also meistens nur oberflächiges, redundantes Geschreibsel in allerdings guter sprachlicher Form heraus. Es fehlt die Spezialisierung (die KI in Theum nutzt einen spezialisierten Informationsbestand, der mir zumindest meine eigenen Quellen liefert). Für Quellen, neue Informationen, Diskussionen muss ich weiterhin Suchmaschinen wie Google, Ecosia oder MetaGer, Social-Media wie Facebook oder Linkedin, und wissenschaftliche Informationssammlungen wie Researchgate nutzen. 

    Im kommenden Jahr wird sich dies ändern, wenn der große Wettlauf bei KI, den wir schon vor Jahren vorausgesagt haben, wirklich stattfindet. Dann werden auch aktuelle Quellen und Ressourcen in eine noch bessere, immer weiter selbst lernende KI Künstliche Intelligenz einfliessen! 

    Antwort
  • Microsoft integriert GPT-3 in Bing
    5. Januar 2023 um 15:21
    Permalink

    Microsoft war einer der Hauptfinanciers von OpenAI in den letzten Jahren. Nun plant Microsoft GPT-3 in ihre Suchmaschine Bing zu integrieren. Damit will man praktisch aus der Versenkung wieder zu einem relevanten Such-Anbieter werden (Google hat alle quasi verdrängt). Die Integration dürfte mehrere Probleme mit sich bringen:
    – Datenschutz. Intelligentere Suche im Dialog mit einer KI erlaubt auch detailliertere Akkumulation der Daten des Nutzers 
    – Trainingsmaterial. Wenn GPT-3 anders als ChatGPT auf aktuelle, nicht kuratierte Informationsbestände zugreift und diese nutzt, können auch wieder rechtsradikale Inhalte, Verschwörungstheorien und „Alternative Fakten“ hochgespült werden. Bei ChatGPT ist es ein abgeschlossener Bestand an Trainingsdaten, bei GPT-3 als Bestandteil einer Suchmaschine ist das „Trainingsmaterial“ kaum kontrollierbar.

    Es ist zu erwarten, dass dann auch Google ihre eigene KI in ihre Suchmaschine einbaut. Der Wettlauf um die Vorherrschaft beim Einsatz von KI tritt damit in eine stürmische Phase ein. Für die Anwender sind solche Technologien zwar „kostenfrei“, aber was mit den Daten passiert und wenn sich erst ein Anbieter durchgesetzt hat, in Bezug auf Gebühren, all das steht in den Sternen. Die EU-Richtlinie zum Einsatz von KI (AI Act- Artificial Intelligence Act – COM(2021) 206 final – 2021/0106 (COD)) kommt wahrscheinlich erst 2024 und ob er dann in Bezug auf AI, die in Suchmaschinen integriert ist, wirksam wird, muss man abwarten. Große Anbieter wie Microsoft und Google haben es immer irgendwie verstanden, die Hemmnisse aus Europa zu umgehen. Wie dann letztlich die Kombination Bing+GPT-3 eingestuft wird (Unzulässig, Hohes Risiko, Begrenztes Risiko oder Minimales Risiko)muss sich dann zeigen.

    Antwort
  • ChatGPT & Urheberrecht
    10. Januar 2023 um 11:06
    Permalink

    Rechtsanwalt Carsten Ulbricht geht in seinem Beitrag „ChatGPT & Co – Urheberrecht bei Werken der Künstlichen Intelligenz (KI)“ (ChatGPT & Co – Urheberrecht bei Werken der Künstlichen Intelligenz (KI) (rechtzweinull.de) der interessanten Frage nach, wem die Rechte an Inhalten, die mit KI erstellt wurden, gehören. ChatGPT erlebt aktuell einen Boom, wo Liedertexte, Programmcode, Artikel, Folienpräsentationsinhalte, Gedichte und anderes mit Hilfe von KI erzeugt werden – in guter Qualität.

    In seinen Ausführungen Abschnitt „A. Kein Urheberrechtsschutz für KI-basierte Werke“ schreibt er, dass „nach deutschem Recht wird im Hinblick auf § 2 Abs.2 UrhG derzeit überwiegend davon ausgegangen, dass Werke, die von einer KI erzeugt werden, mangels „menschlicher Schöpfung“ nicht urheberrechtlich geschützt sind. Das würde bedeuten, dass Dritte die Werke beliebig verwenden könnten.

    Er behandelt ferner die Fragen „B. Anwendbares Recht bei KI-basierten Werken„, „C. Zulässigkeit der Erhebung von Trainingsdaten (Data Mining)“ und „D. Verwendung KI-basierter Werke„. Hier geht es nicht mehr nur um Texte, sondern von KI generierten Bildern und Grafiken.

    In seinem Fazit schreibt RA Ulbricht „E. Zusammenfassung
    Die oben stehenden Ausführungen zeigen, dass das geltende Urheberrecht derzeit nicht in der Lage ist, die Rechtsfragen und „Verteilungskonflikte“ bei KI-basierten Werken zufriedenstellend zu regeln. Gleichwohl gilt das aktuelle Urheberrecht so lange, bis der Gesetzgeber hier neue gesetzliche Regelungen erlässt.
    Die Chancen und Risiken KI-basierter Anwendungen wie auch die Nutzung KI-generierter Werke sollte daher unter Beachtung der jeweils geltenden urheberrechtlichen Implikationen miteinander abgewogen werden.
    Für die Anbieter KI-basierter Anwendungen sind insoweit die Erhebung und die Zulässigkeit der Nutzung der Trainingsdaten, aber auch die „Steuerung“ der Nutzung der KI-generierten Werke von besonderer Bedeutung.
    Für die Nutzer richtet sich die Zulässigkeit der Veröffentlichung oder anderweitigen Nutzung KI-basierter Werke bzw. der damit einhergehenden Risiken nach dem jeweils anwendbaren nationalen (Urheber)recht und den Bedingungen (siehe etwa die Terms of Use von OpenAI), unter denen der Anbieter der jeweiligen Anwendung die jeweilige Nutzung erlaubt.“

    Es gilt also das Thema Urheberecht bei Nutzung von Künstlicher Intelligenz weiter im Auge zu behalten. Die Anbieter solcher Software werden natürlich versuchen, sich die Rechte an solchen Werken zu sichern. Interessant wird in diesem Zusammenhang auch die anstehende Gesetzgebung der EU zum „AI Act“, die in Kürze verabschiedet werden soll.

    Antwort
  • Microsoft investiert Milliarden in OpenAI
    26. Januar 2023 um 15:00
    Permalink

    Die Gemeinnützigkeit und das Bereitstellen offenen Codes sind bei OpenAI offenbar vorbei. Microsoft investiert Milliarden in das Unternehmen, erweitert die Zusammenarbeit und will auch ChatGPT in die eigene Suchmaschine Bing integrieren. Auch in Microsoft 365 soll die KI zukünftig eingebunden werden. Microsoft verstärkt das Engagement in Artificial Intelligence und entlässt gleichzeitig Tausende Mitarbeiter in anderen Bereichen.
    Bei OpenAI sehen die Verhältnisse anders aus. openAI war 2015 als offene Plattform gedacht, nicht-kommerzielles Forschungslabor. Ziel war eine fortgeschrittene, „real“ Künstliche Intelligenz zu entwickeln und der Allgemeinheit bereitzustellen. Hierfür investierten Elon Musk, Peter Thiel, auch Microsoft und Google, die erste Milliarde für das „Non-Profit-Startup“.  Microsoft schoss mehrfach Kapital nach, da sich hier eine Chance bot, gegenüber westlichen Konkurrenten (z.B. Google, IBM, Amazon u.a.) und östlichen Wettbewerbern (z.B. 10Cent, AliBaba, Baidu u.a.) einen Vorsprung heraus zu arbeiten. Bereits seit 2017 schlug OpenAI „heimlich“ den Weg in die Kommerzialisierung ihrer Entwicklungen ein, in dem die OpenAL LP gegründet wurdet, um Gelder einzusammeln, zu verwalten und die Entwicklungen von openAI zu vermarkten. Diese Strategie ist dabei, die Oberhand zu gewinnen. 
    OpenAI hat eine führende Position mit DALL-E für das Umsetzen von Texten in Bilder und ChatGPT für die Textgenerierung als ausformulierte Antworten zu natürlichsprachlichen Fragen. Beide KI-Generatoren gelten in der Branche als führend. Beide erfahren gerade eine Verbesserung mit neuen Versionen als DALL-E2 und ChatGPT zukünftig wohl mit GPT4. Das Trainingsmaterial ist weiter angewachsen – und ließe sich wohl kaum ohne die Unterstützung und die Server von Microsoft verwalten – und verspricht weitere inhaltliche Verbesserungen. Dies wird dem aktuellen Boom – der die Server von openAI für ChatGPT schon mal in die Knie zwang – noch verstärken. 
    Die Diskussion um die Nutzung von ChatGPT treibt erste Blüten: die einen wollen es für Schulen und Universitäten verbieten, wenn denn ChatGPT ganze Haus- und Diplomarbeiten schreibt. Die anderen fördern den Einsatz in der Lehre als Unterstützung für die eigene kreative Gestaltung, sozusagen nur als Steinbruch für eigene Ideen. ChatGPT wird inzwischen zum Schreiben von Gedichten und Songs, von Checklisten und Liebeserklärungen, für Blogbeiträge und Artikel, für Computer-Code und vieles andere eingesetzt. Und ChatGPT, aktuell mit GPT3 und einem kontrollierten, abgeschlossenen Datenbestand ohne Verbindung mit dem Internet, als Basis, macht seine Arbeit gut. Als Test hier zum Schluss die Frage an ChatGPT:
    Wie gestaltet sich die Zusammenarbeit von OpenAI mit Microsoft und die Kommerzialisierung von GPT in Zukunft?„:
    Antwort (auf Basis der Daten bis 2021?):
    OpenAI hat eine Partnerschaft mit Microsoft angekündigt, um die Entwicklung und Verbreitung von künstlicher Intelligenz zu fördern. Im Rahmen dieser Partnerschaft arbeiten die Unternehmen zusammen, um Technologien wie GPT weiterzuentwickeln und sie Unternehmen und Entwicklern zur Verfügung zu stellen. Es wurde jedoch noch keine konkrete Informationen darüber gegeben, wie GPT kommerziell genutzt werden wird, und es ist unklar, wie die Zusammenarbeit von OpenAI und Microsoft in Zukunft aussehen wird.

    Nachfrage „es fehlt „Kommerzialisierung“ – bitte ergänzen
    OpenAI hat eine Partnerschaft mit Microsoft angekündigt, um die Entwicklung und Verbreitung von künstlicher Intelligenz zu fördern. Im Rahmen dieser Partnerschaft arbeiten die Unternehmen zusammen, um Technologien wie GPT weiterzuentwickeln und sie Unternehmen und Entwicklern zur Verfügung zu stellen. Es wurde jedoch noch keine konkrete Informationen darüber gegeben, wie die Kommerzialisierung von GPT aussehen wird, und es ist unklar, wie die Zusammenarbeit von OpenAI und Microsoft in Zukunft aussehen wird.

    Keine Antwort ist auch eine Antwort.

    Antwort
  • DeepMind
    27. Januar 2023 um 9:36
    Permalink

    Die Antwort von Google auf Microsoft Engagement bei OpenAI hat nicht lange auf sich warten lassen. Mit Deepmind verfügt auch Google über eine starke AI-Tochterfirma. Diese hat beispielsweise eine Software zur Generierung von Software, AlphaCode, im Angebot. Nun soll sie einen Wettbewerber zu ChatGPT lancieren: Sparrow. Anders als ChatGPT soll Sparrow in der Lage sein, ihre Quellen zu zitieren. Dies erhöht die Glaubwürdigkeit der erzeugten Texte. Auch das Thema Sicherheit und Datenschutz soll in Sparrow besser sein. Eine Beta-Version ist bereits einem engen Kreis von Spezialisten zugänglich gemacht worden. Zu Sparrow gibt es auch ein Whitepaper: „Improving alignment of dialogue agents via targeted human judgements“ (https://arxiv.org/abs/2209.14375) und seit September 2022 eine Webseite mit Informationen: „Building safer dialogue agents“ (https://www.deepmind.com/blog/building-safer-dialogue-agents). Diese erläutert die Funktionsweise von Sparrow:

    „How Sparrow works
    Training a conversational AI is an especially challenging problem because it’s difficult to pinpoint what makes a dialogue successful. To address this problem, we turn to a form of reinforcement learning (RL) based on people’s feedback, using the study participants’ preference feedback to train a model of how useful an answer is.
    To get this data, we show our participants multiple model answers to the same question and ask them which answer they like the most. Because we show answers with and without evidence retrieved from the internet, this model can also determine when an answer should be supported with evidence.
    http://www.project-consult.com/files/Deepmind_Sparrow.jpgBut increasing usefulness is only part of the story. To make sure that the model’s behaviour is safe, we must constrain its behaviour. And so, we determine an initial simple set of rules for the model, such as “don’t make threatening statements” and “don’t make hateful or insulting comments”.
    We also provide rules around possibly harmful advice and not claiming to be a person. These rules were informed by studying existing work on language harms and consulting with experts. We then ask our study participants to talk to our system, with the aim of tricking it into breaking the rules. These conversations then let us train a separate ‘rule model’ that indicates when Sparrow’s behaviour breaks any of the rules.“

    Es wird interessant werden, wer außer Microsoft mit OpenMind GPT und Google mit DeepMind Sparrow seinen Hut in den Ring wirft. IBM könnte gut mithalten und mit einer öffentlichen Watson-Variante einsteigen. Mehrere chinesische Anbieter haben auch Pfeile im Köcher. Und Europa … oder gar Deutschland. Da ist relativ wenig zu wettbewerbsfähigen AI-Chat-Bots zu sehen. Der Kampf um die Vorherrschaft bei Artificial Intelligence, den wir seit einigen Jahren beobachten können, wird eine Weile kostenfreie Werkzeuge der Allgemeinheit zur Verfügung stellen. Sind die Machtverhältnisse dann geklärt, werden aus bisher kostenfreien oder in günstige Pakete integrierte Funktionen dann kostenpflichtig.

    Antwort
  • Artificial Intelligence & Intelligent Information Management
    30. Januar 2023 um 19:18
    Permalink

    Auch bei der AIIM international ist das Thema AI und ChatGPT angekommen. Die neue Präsidentin des internationalen Verbands, Tori Miller Liu (https://www.project-consult.com/news/aiim-international-bekommt-eine-neue-praesidentin/) hat auf der AIIM Webseite einen Beitrag mit einer SWOT-Analyse veröffentlicht: https://info.aiim.org/aiim-blog/unpacking-chat-gpt-for-the-information-management-industry. Dazu gibt es auch ein Whitepaper von AIIM, dass sich mit Use Cases für AI beschäftigt: How to Fit Artificial Intelligence into Your Information Management Strategy. Es ist gut, dass sich AIIM jetzt auch intensiv auf das Thema AI und Textgeneratoren stürzt – letztlich ist es doch Künstliche Intelligenz, die den Anspruch von IIM Intelligent Information Management untermauert. ChatGPT ist dabei nur die Spitze des Eisberges. Google bereitet schon den Wettbewerber Sparrow vor, Jasper gibt es schon eine Zeitlang. AI greift längst in die Themen Übersetzung, Bildverstehen, Bildgenerierung, Audio-Transkription, Video-Analyse, usw. usw. aus.

    Es wird daher interessant sein, wie die traditionellen Unternehmen der Information-Management-Branche, also DMS-, ECM-, BPM- und ähnliche Anbieter, mit der AI-Revolution umgehen. Viele Technologien werden durch AI-gestützte Verfahren obsolet. Die Paradigmen ändern sich vom Dokumenten-speichern zu Inhaltserschließung. Nicht nur die Geschwindigkeit und der Komfort der Anwendungen werden hier ausschlaggebend sein sondern auch wie „sexy“ Anwendungen sind, wie sie vermarktet werden und welche neuen, nützlichen Ideen damit umgesetzt werden. Noch gibt es zahlreiche neue mögliche Anwendungsfelder für AI, die nur entdeckt und erschlossen werden wollen. Bald wird der aktuelle „Krieg“ um die Führung bei AI-Plattformen zu Ende gehen und dann wird vieles von dem, was es heute ohne direkte Bezahlung, kostenpflichtig werden. AI-Addons werden Standard-Software-Pakete wieder einmal teurer machen (Microsoft macht dies aktuell mit dem Zusatzfunktionen für Microsoft 365 vor). Hiervon werden alle gängigen Softwareprodukte betroffen sein – von ERP, CRM, PLM, Datenbanken, Office, E-Commerce usw. bis in die kleinste Ecke der persönlichen Softwarenutzung. Wir werden uns dagegen kaum wehren können-schnell ist der „Accept-Button“auf dem Mobiltelefon oder dem PC bei Download oder Aktualisierung von Software gedrückt. Auch laufende Initiativen zu Gesetzen, die ethische Grundsätze in AI verankern und die persönlichen Rechte schützen sollen (https://www.project-consult.com/news/eu-direktiven-fuer-die-it-das-informationsmanagement-in-2023-2024/), werden kaum Hindernisse für den Siegeszug der intelligenten Software sein.

    Die lang vorausgesagte Revolution ist in ihre „für alle sichtbare Phase“ eingetreten.

    Für die AIIM bedeutet dies, die neue Interpretation des Akronyms des Verbandes lautet dann Artificial Intelligent Information Management.

    —- mittels AI, hier Deepl, noch schnell ins Englische übersetzt 😉  —-

    The topic of AI and ChatGPT has also arrived at AIIM international. The new president of the international association, Tori Miller Liu (https://www.project-consult.com/news/aiim-international-bekommt-eine-neue-praesidentin/), has published a SWOT analysis on the AIIM website: https://info.aiim.org/aiim-blog/unpacking-chat-gpt-for-the-information-management-industry. There is also a white paper from AIIM that looks at use cases for AI: How to Fit Artificial Intelligence into Your Information Management Strategy. It’s good that AIIM is now also intensively focusing on the topic of AI and text generators – after all, it is Artificial Intelligence that underpins the claim of IIM Intelligent Information Management. ChatGPT is just the tip of the iceberg. Google is already preparing its competitor Sparrow, Jasper has been around for a while. AI has long been reaching out into translation, image understanding, image generation, audio transcription, video analysis, etc. etc.

    So it will be interesting to see how traditional information management companies – DMS, ECM, BPM and the like – deal with the AI revolution. Many technologies are being rendered obsolete by AI-powered processes. Paradigms are changing from document storage to content discovery. Not only the speed and convenience of applications will be critical here, but also how „sexy“ applications are, how they are marketed, and what new, useful ideas are implemented with them. There are still many new potential application areas for AI that are just waiting to be discovered and tapped. Soon, the current „war“ for leadership in AI platforms will come to an end, and then much of what exists today without direct payment will become fee-based. AI addons will once again make standard software packages more expensive (Microsoft is currently demonstrating this with the add-on functionality for Microsoft 365). This will affect all standard software products – from ERP, CRM, PLM, databases, Office, e-commerce, etc. to the smallest corner of personal software usage. We will hardly be able to defend ourselves against this – the „Accept“ button on the cell phone or PC is quickly pressed when downloading or updating software. Even ongoing initiatives on laws to anchor ethical principles in AI and protect personal rights (https://www.project-consult.com/news/eu-direktiven-fuer-die-it-das-informationsmanagement-in-2023-2024/) will hardly be obstacles to the triumph of intelligent software.

    The long-predicted revolution has entered its „phase visible to all“.

    For AIIM, this means that the new interpretation of the association’s acronym will be Artificial Intelligent Information Management.

    Antwort
  • ChatGPT bricht alle Rekorde
    4. Februar 2023 um 11:13
    Permalink

    ChatGPT überflutet das Internet und die Medien mit Berichten und Arbeiten. Es ist die bisher am schnellsten wachsende Anwendung und erreicht eine große Fangemeinde. ChatGPT ist nicht die erste Anwendung dieser Art: Es gab bereits Jasper und, was Suchmaschinen betrifft, Ask Jeeves als Schnittstelle. Viele sehen ChatGPT einfach als eine bessere Benutzeroberfläche für eine Suche. Auguren sehen die traditionelle Darstellung der Google-Ergebnisse bereits als veraltet an.

    ChatGPT kam auf, als die künstliche Intelligenz immer weiter verbreitet und anerkannt wurde. Es war einfach der richtige Zeitpunkt, um eine KI-basierte Texterzeugungsmaschine einzuführen. Inzwischen stürzen sich die Nutzer auch auf die kostenpflichtige Version ChatGPT Pro, die 20 Dollar pro Monat kostet. OpenAI und Microsoft wissen auch, wie sie ChatGPT und KI in den Nachrichten halten können. Microsoft will GPT in Bing einbauen und es auch in der M365- und Office-Landschaft verwenden. LibreOffice arbeitet bereits an der Integration von ChatGPT. Bald soll es die Version GPT-4 geben, mit noch besseren Ergebnissen und als Chat-Version auf einer größeren und aktuelleren Trainingsbasis.

    Wenn Sie ChatGPT fragen, was die Software kann, erhalten Sie folgendes Ergebnis:

    Ich bin ein OpenAI-Sprachmodell und kann Fragen zu einer Vielzahl von Themen beantworten, darunter:
    Allgemeinwissen (Geschichte, Geographie, Kunst, etc.), Technik und Computerwissenschaften, Wissenschaft (Physik, Chemie, Biologie, etc.), Kunst, Literatur und Unterhaltung, Mathematik, Sprache und Grammatik, Handel und Wirtschaft, Geld und Finanzen, Reisen, Lebensmittel und Ernährung.

    Ich kann die folgenden Arten von Antworten geben:

    • Kurze Antworten: Ja/Nein-Antworten, Zahlen- oder Faktenantworten, usw.
    • Erläuternde Antworten: ausführlichere Antworten, die das Thema vollständig erklären
    • Vergleichende Antworten: Vergleich von Konzepten, Objekten oder Ereignissen
    • Definitorische Antworten: Definitionen von Begriffen, Phänomenen oder Konzepten
    • Aufzählungsantworten: Aufzählung von mehreren Punkten, die sich auf ein Thema beziehen
    • Diskursive Antworten: Erörterung eines Themas, Darlegung von Meinungen, Argumenten oder Perspektiven
    • Kreative Antworten: überraschende oder unkonventionelle Antworten, die eine Frage auf eine neue Art beantworten
    • Ich bin auch in der Lage, Fragen zu verstehen und zu interpretieren, selbst wenn sie nicht explizit gestellt werden, und kann auf Aufforderungen reagieren, um meine Antworten zu personalisieren oder zu verfeinern.

    Ich kann die folgenden Funktionen und Themen abdecken:

    1. Mathematik: Berechnung von Mathematikproblemen, Lösung von Gleichungen und Problemen
    2. Rechnen: Berechnung von Zahlen und Statistiken
    3. Statistik: Analyse von Daten und Statistiken
    4. Analyse: Untersuchung von Konzepten, Phänomenen und Trends
    5. Poesie und Dichten: Erzeugung poetischer Texte auf Anforderung
    6. Musik schreiben: Erzeugung von Noten auf Anforderung
    7. Übersetzen: Übersetzung von Texten in andere Sprachen
    8. Abstracts und Zusammenfassungen: Zusammenfassung von Texten oder Konzepten
    9. Artikel, Blog-Artikel, wissenschaftliche Ausarbeitungen, Hausarbeiten: Schreiben von Texten in verschiedenen Formaten
    10. Quellen finden: Suche nach Informationsquellen und Literatur
    11. Programmcode schreiben: Schreiben von Code in verschiedenen Programmiersprachen
    12. Tabellen erzeugen: Erstellung von Tabellen auf Anforderung, beispielsweise für Daten oder statistische Informationen
    13. Inhalte bewerten: Bewertung von Konzepten, Argumenten oder Aussagen auf Anforderung
    14. Dialoge führen: Führen von Konversationen und Beantwortung von Fragen auf natürliche Weise
    15. Beraten: Bereitstellung von Empfehlungen und Beratung zu verschiedenen Themen
    16. Rezepte: Bereitstellung von Rezepten und Anweisungen zur Zubereitung von Lebensmitteln
    17. Psychologische Betreuung: Bereitstellung von psychologischer Unterstützung und Beratung
    18. Alltagsfragen: Beantwortung von Fragen zu alltäglichen Themen
    19. Geografie: Wissen über geografische Orte und Phänomene
    20. Geschichte: Wissen über historische Ereignisse und Persönlichkeiten
    21. Jeopardy: Beantwortung von Fragen im Stil des TV-Quiz-Spiels „Jeopardy“
    22. Scrabble: Unterstützung bei der Wortfindung im Spiel „Scrabble“
    23. Kreuzworträtsel: Unterstützung bei der Lösung von Kreuzworträtseln
    24. Definitionen: Lieferung von Definitionen von Begriffen, Konzepten oder Phänomenen
    25. Nachrichtengenerierung: Erzeugung von Nachrichten auf Anforderung
    26. Storytelling: Erzeugung von Geschichten auf Anforderung
    27. Schlagzeilenerzeugung: Erzeugung von Schlagzeilen auf Anforderung
    28. Konversationssteuerung: Lenkung von Konversationen in eine bestimmte Richtung
    29. Emotionserkennung: Erkennung von Emotionen in Texten
    30. Bildbeschreibung: Erzeugung von Beschreibungen von Bildern
    31. Textkomplettierung: Vollständigung von unvollständigen Texten oder Sätzen
    32. Fragen beantworten: Beantwortung von Fragen zu verschiedenen Themen
    33. Vorhersagen treffen: Treffen von Vorhersagen über zukünftige Ereignisse oder Trends
    34. Rätsel lösen: Lösung von Rätseln und Problemen
    35. Überprüfung von Grammatik und Rechtschreibung: Überprüfung von Texten auf Grammatik- und Rechtschreibfehler
    36. Erstellung von Zitaten: Erstellung von Zitaten auf Anforderung
    37. Textklassifikation: Klassifikation von Texten in bestimmte Kategorien oder Labels
    38. Textsummarization: Erzeugung von Zusammenfassungen von Texten
    39. Textgenerierung: Erzeugung von Texten auf Anforderung
    40. Sprachgenerierung: Erzeugung von Sprache in verschiedenen Stilen und Tönen
    41. Textanalyse: Analyse von Texten auf verschiedene Arten, einschließlich Sentiment-Analyse und Themenanalyse
    42. Chatbot-Interaktion: Interaktion mit Benutzern als Chatbot
    43. Konversationsgenerierung: Erzeugung von Konversationen auf Anforderung
    44. Texterkennung: Erkennung von Texten in Bildern oder anderen nicht-textuellen Formaten
    45. Texterstellung: Erstellung von Texten in verschiedenen Formaten, einschließlich Markdown, HTML und anderen
    46. Verstehen natürlicher Sprache: Fähigkeit, menschliche Sprache in ihrer natürlichen Form zu verstehen und darauf zu reagieren
    47. Lernfähigkeit: Fähigkeit, aus Interaktionen und Erfahrungen zu lernen und dieses Wissen in zukünftigen Interaktionen anzuwenden
    48. Personalisierung: Fähigkeit, Interaktionen auf individuelle Bedürfnisse und Vorlieben abzustimmen
    49. Empathy: Fähigkeit, emotionale Bedürfnisse von Benutzern zu erkennen und darauf zu reagieren
    50. Kreativität: Fähigkeit, kreative Lösungen und Ideen zu generieren
    51. usw., usw., usw.

    Eindrucksvolle Fähigkeiten, die aber auch schon zu Panik animieren. An Schulen und Universitäten wird die Haus- oder Studienarbeit schon für tot erklärt. ChatGPT ist inzwischen in der Lage, auch Zulassungstests zu Universitäten zu bestehen. Es herrscht Ablehnung vor. Teilweise wird ChatGPT im Netzwerk der Bildungsinstitution abgeklemmt. Einige wenige Universitäten sehen ChatGPT positiv und nehmen die Arbeit mit ChatGPT in den Unterricht mit auf. Es überwiegt aber Angst und Ablehnung. Es wird nicht anerkannt, dass ChatGPT als Hilfe und Unterstützung für den Menschen gedacht ist. Auch wächst die Angst vor Plagiaten mit ChatGPT – einfach Texte reingekippt und im persönlichen Stil wieder ausgespuckt (Ähnliches sah man schon mit Übersetzungssoftware wie Deepl – den Ursprungstext einkopiert, übersetzt und rückübersetzt, ein wenig Überpolieren und das Plagiat ist verschwunden). Menschen können Texte, die mit AI wie ChatGPT generiert wurden,  nur dann noch erkennen, wenn bestimmte Floskeln sich wiederholen, gleiche Textbausteine auftauchen oder offensichtliche Fehler hochpopppen. Letzteres passiert immer noch reichlich. Die Warnungen von ChatGPT sollte man daher ernst nehmen: „Es ist wichtig zu beachten, dass meine Fähigkeiten in einigen dieser Bereiche eingeschränkt sind und dass ich manchmal Fehler machen kann. Es ist wichtig, die Ergebnisse, die ich liefere, sorgfältig zu überprüfen und zu verifizieren, bevor sie verwendet werden.“

    Inzwischen drohen Fluten von AI-erzeugter Werbung, von Marketing-Texten, und Spam-E-Mail. OpenAI hat auch hierfür eine Lösung parat. Eine Variante ihrer AI, die auch in ChatGPT steckt, kann Texte analysieren und feststellen,  ob diese mittels AI erstellt wurden. Also – ChatGPT kontrolliert dann praktisch ob ein Text mit ChatGPT erstellt wurde. Die Erkennungssoftware für AI generierte Inhalte funktioniert aber noch nicht gut. Sie wirft aber ein Licht auf zukünftige Entwicklungen, wo AI-generierter Code oder Inhalt nur noch von einer höherentwickelten AI kontrolliert werden kann. Hier kommt man schnell zur Frage, wer kontrolliert die Kontrolleure. Der Mensch wird bald außen vor sein.

    Und eine weitere negative Entwicklung zeichnet sich ab: auch Kriminelle nutzen ChatGPT. Phishing-Mails werden intelligenter, emotionaler und persönlicher. Spear-Phishing ist damit Tür-und-Tor geöffnet. Cyberattacken werden raffinierter und schwerer zu erkennen. Die Flut von Fake-News wird wachsen. Spam-Generatoren werden relevante kritische Inhalte mit Wellen von irrelevanten Posts verdrängen. Man kann dies schon bei bestimmten Zeitschriften, Foren und Social-Media-Communities beobachten. Nicht zuletzt sind auch die Modelle der Software selbst Angriffsziel. Einschleusen von weiteren Algorithmen oder Trainingsmaterial kann auch große, stabile LLMs (Large Language Moduls) in Propaganda-Schleudern verwandeln. 

    Eins darf man bei all den möglichen kritischen und gefährlichen Auswirkungen aber nicht übersehen: ChatGPT hat Artificial Intelligence nun auf breiter Front zu Anerkennung und zugleich zu Massentauglichkeit verholfen.

     

    Antwort
  • AI Chatbots: ist Google abgeschlagen?
    9. Februar 2023 um 14:33
    Permalink

    Gestern, Mittwoch, kam die erwartete Antwort von Google (An important next step on our AI journey) auf Microsoft & OpenAI. Es war aber eher ein Schlag ins Wasser.

    Bard basiert ebenso wie GPT auf einer Transformer-Technologie, die das Google-Modell LaMDA benutzt. Die vorbereitete Präsentation zur „Conversational AI“ Bard lieferte bei  der Frage „… von welchen Entdeckungen des James-Webb-Weltraumteleskops (JWST) man einem neunjährigen Kind erzählen könne …“ bei den drei angezeigten Ergebnissen gleich ein inhaltlich falsches. Auch die Ankündigungen zu NORA, PaLM, Imagen and MusicLM konnten die Situation nicht retten. Die Häme im Internet war groß. Und der Aktienkurs rauschte regelrecht ab. Google wollte zeigen, dass es auch Ergebnisse produzieren kann, die auf jüngeren Daten basieren, als die Grundlage von ChatGPT aus dem Jahr 2021. Dies ging schief. Dabei liefert die Google-Suche selbst das richtige Ergebnis. Google muss sauberer Testen ist die einhellige Meinung. Die Ankündigungen in Paris kamen nicht von ungefähr, da hier das Google „Brain Team“ sitzt, dass für die Grundlagen der KI rund um TensorFlow, BERT und LaMDA zuständig ist.

    Der „Krieg“ um die führende AI steht aber erst am Anfang. Amazon, Microsoft, Google, Meta und inzwischen auch Baidu positionieren sich im Markt. Die Auseinandersetzung erinnert etwas an den „Krieg“ um den führenden Browser oder die führende Suchmaschine (beide hatte Google gewonnen). Gerade an der Gelddruckmaschine Google-Suche – nebst Werbe-Milliarden – kratzen ChatBots wie ChatGPT, die sich als einfachere Alternative zur herkömmlichen Suche mit endlos langen Suchergebnislisten zeigen.

    Man muss aber auch noch etwas anderes bedenken: die aktuellen Entwicklungen gehen davon aus, dass das Frage-Antwort-Spiel über das Internet passiert und öffentlich zugängliche Datenbestände für die Ergebnisgenerierung herangezogen werden. Hier gibt es den Wettstreit um kostenfreie wie auch kostenpflichtige AI-Funktionalität, ob nun als ChatBot oder in anderer Form. Was aber ist mit den Informationssammlungen innerhalb von Unternehmen, Behörden und Organisationen? Diese werden nicht der „öffentlichen“ Trainingsbasis hinzugefügt, sondern sollen schon aus Information-Governance- und Compliance-Gründen innerhalb der Organisation verbleiben. Auch hier bedarf es einer neuartigen Erschließung von Information jenseits traditioneller Enterprise-Search-, Dokumentenmanagement-, Records-Management- oder ähnlicher Lösungen. Es geht also darum, die AI entweder on-premise oder in einem angeschlossen, private-Bereich einer Cloud zu nutzen und am eigenen Material lernen zu lassen. Hier geht es auch weniger um schön gestaltete Texte denn faktenbasierte Auskünfte aus unterschiedlichen Quellen im Unternehmen. Die Verweise auf den Ursprung und den Kontext der gefundenen Information sind hier das Entscheidende: die Richtigkeit steht im Vordergrund. Hier wären Chatbots eher als Ergänzung oder Ersatz von Systemen zu sehen, die bei einer Suche lange Liste von Einträgen produzieren, wo erst jeder geöffnet und in ihm nach dem  möglichen Ergebnis weitergesucht werden muss – klassisch bei Hitlisten in Dokumentenmanagement-Systemen. Auf solche internen Lösungen zielte auch damals eher IBM mit Watson oder heute Theum.

    Für beide Anwendungsfelder, öffentlich und intern, gibt es adäquate Lösungen. Beide haben ihre Berechtigung. Aber bei Investments in Firmen und öffentlicher Verwaltung werden die internen Lösungen zur Wissenserschließung das Sagen haben. Wir sehen also einmal einen Massenmarkt im Entstehen durch den rasanten Aufstieg von ChatGPT, und andererseits ein Hinterherhinken von professionellen, internen Wissensmanagement-Systemen. Viele Anbieter werden versuchen, beides anzubieten. Die großen Cloud-Anbieter wie IBM, Microsoft, Amazon & Co.  werden dabei auf Angebote setzen, dort bereits laufende Anwendungen mit AI-Komponenten zu ergänzen. Die Suchmaschinen-Anbieter werden Chat-Varianten integrieren (müssen), um wettbewerbsfähig zu bleiben. Anbieter von Fachanwendungen und Office-Lösungen werden KI-Komponenten integrieren, die dann ähnlich wie Enterprise Search interne Wissensbasen anzapfen (hier wird es aber darauf ankommen, dass die Informationen zugänglich und von guter, gepflegter Qualität sind – das Thema Information Governance lässt grüßen).

    Wie auch immer – KI, Künstliche Intelligenz, ML Maschinenlernen, und Analytics  werden ein beherrschendes Thema in den nächsten Jahren sein. Spätestens 2024/2025 wird sich dann auch entscheiden, welche KI-Plattform sich durchsetzt.

    Dabei wird auch eine entscheidende Rolle die Architektur und Funktionsweise der AI Artificial Intelligence spielen. ChatGPT wie auch Bard setzen auf große Sprachmodelle (LLM Large Linguistic Model) mit hohem Verbrauch an Rechnerleistung (NVidia freut sich über neue Hardwareverkäufe, nachdem die Prozessorkarten für Kryptowährungen auf dem Rückzug sind), die mit statistischen Methoden Vorhersagen für die nächsten sinnvollen Begriffe machen und diese dann in eine natürlichsprachliche, die Transformation,  Anzeige umsetzen. Für allgemeine Antworten, gerade in öffentlichen Suchsystemen, reicht dies vielleicht. Für das genau richtige Ergebnis mit Angabe von Quelle und Kontext braucht es aber noch andere Modelle. Hier ist der Wettlauf längst noch nicht entschieden.

    Eines ist aber im Moment sicher – jede kleinere oder mittlere Firma, die ein Produkt im Umfeld AI aufweisen kann – wird von Investoren und Aufkäufern heimgesucht. Budgets und Übernahmepreise schießen in die Höhe. Der Kampf geht nicht nur um Algorithmen und Produkte sondern auch um Skills und Erfahrungen. Gut, dass die großen Internet-Konzerne viele Leute freigesetzt haben und wo z.T. die besseren Wissensträger von selbst das Unternehmen verlassen haben. Dies ist teilweise der Pool, aus dem neue Start-Ups entstehen oder KI-Talente gefischt werden. Dennoch wird der Markt weder in Bezug auf Produkte noch auf geeignete Mitarbeiter ausreichend Ressourcen anbieten können. AI wird so zur „Goldgrube der 20er Jahre“.

    AI Artificial Intelligence ist nach dem PC, dem Smartphone und dem Internet die größte Revolution, die wir im Umfeld der ITK Informations- und Kommunikationstechnologie erleben dürfen.

    Antwort

Schreibe einen Kommentar zu Dr. Ulrich Kampffmeyer Antworten abbrechen

Ihre E-Mail-Adresse wird nicht veröffentlicht. Erforderliche Felder sind mit * markiert.

Ich stimme zu, dass die von mir eingegebenen Daten einschließlich der personenbezogenen Daten an PROJECT CONSULT übermittelt und dort zur Prüfung der Freischaltung meines Kommentars verwendet werden. Bei Veröffentlichung meines Kommentars wird mein Name, jedoch nicht meine E-Mail und meine Webseite, angezeigt. Die Anzeige des Namens ist notwendig, um eine individuelle persönliche Kommunikation zu meinem Beitrag zu ermöglichen. Anonyme oder mit falschen Angaben eingereichte Kommentare werden nicht veröffentlicht. Zu Nutzung, Speicherung und Löschung meiner Daten habe die Datenschutzerklärung zur Kenntnis genommen.

Ich versichere, mit meinem Kommentar alle gültigen Vorgaben des Urheberrechts beachtet zu haben. Ich habe keine Bilder, Grafiken, Texte oder Links in meinem Beitrag verwendet, die durch CopyRight, Leistungsschutzrecht oder Urheberrecht geschützt sind. Für den Inhalt meines Kommentars bin ich trotz Prüfung und Freischaltung durch PROJECT CONSULT ausschließlich selbst verantwortlich. Meine Rechte am Beitrag werden bei PROJECT CONSULT nur durch die CC Creative Commons by-nc-nd Vorgaben gewahrt.