The flip side of generative AI: Extractive AI

Extractive AI takes a more comprehensive and transparent approach to machine intelligence.

Should we go back to paper-based KM?

The sheer volume of largely useless data we have accumulated across the years severely limits the ability of AI to work well, and it comes at a heavy environmental and financial cost.

The trust problem with GenAI

2023 has been the year of ultra-hyping GenAI, and who is paying for this deluge of marketing? Technology vendors that want us to buy it. Again, it's impressive stuff, but when we shift from selling to buying and ultimately using it, many tough questions need to be asked.

8 billion and counting

The message is clear: No single person or committee or group can weave the best paths through the infinite maze of possible event chains. Only humans and machines working together, side by side, can produce a better result than would ever be possible from either one alone.

The five ages of data

Perhaps this latest phase in the history of data will bring us to accept inexplicable complexity as a property of the world. We could view this as pure chaos, but thanks to having lived through the past four ages in rapid succession, we might instead recognize that chaos as being rich with endless mysteries we will never uncover completely.

Truth, lies, and large language models

The good news is that the problem of chat AI's proclivity for hallucinating is well-recognized by the organizations creating these marvels, and they realize that it is a danger to the world and to their success, not necessarily in that order of priority. Until that problem is solved, chat AI engines need to lose their self-confidence and make it crystal clear that they are the most unabashed and charming liars the world has ever seen.

When is good enough enough?

Our goal should be to improve the quality of knowledge assets and their accuracy and relevance in use. Much of this will come from human expertise and effort, increasingly combined with the power of AI.

Get your game on: KM skills needed for reliable use of LLMs

There is no questioning that generative AI is here to stay, but its use in mission-critical work has some way to go before it can be trusted and let loose.

What are your chatbot’s pronouns?

We don't have pronouns by which we can address inanimate objects because we haven't had any occasions to have actual conversations with them.

Are you data-driven or knowledge-driven?

We no longer need to blindly accept the output of even the most sophisticated AI/ML platforms. In fact, we should not consider any artifact, whether produced by humans or machines, as valid knowledge unless it contains not only supporting data and analyses, including provenance, but also an explanation of the underlying plausibility.

The ChatGPT ways of knowledge

These two types of knowing—understanding the world and understanding knowledge—are, in some important ways, at odds in AI-based chatbots.

Tags, AI, and dimensions

Tags have become so common that they've faded from consciousness since 2007, although sometimes a clever hashtag pops up.

Look to the skies for KM opportunities

Then there's the inevitable demand for more automation, from the flight planning and clearance process to the operation of the air vehicles themselves. No human or group of humans could possibly keep track of so many constantly changing variables

AI technologies upending traditional KM

If we are not careful and proactive about it, the concept and importance of knowledge itself may soon become blurred or lost.

The effect of ChatGPT on KM

At this peak of ChatGPT hype, we have to ask what value it may bring.

AI’s new type of knowledge

This way of knowing works pragmatically for some very complex systems of the sort we find in the real world. But, oddly, itseems not to work so well in some artificially simple systems.

Introducing work intelligence systems

Technological advances are significant and can bring huge benefits, but only as long as you understand that they can advise, augment, and support, but not replace, you.

The human capability to under-or overestimate

Yet maybe the most glaring example of underestimating humans we encounter in our work is in the world of AI. It's partly the term "intelligence" in AI that misleads so many, as AI is not intelligent in the same way that humans are intelligent. Though powerful, AI ultimately matches patterns it has learned, and even the smartest of AI systems is limited in how many patterns it can match and make sense of.

The undiscovered country

Capturing and sharing what you already know is good; and with today's data and text analytics tools, it has become much easier than when we'd first begun this journey.

What ‘sentient’ AI teaches us

As Gary Marcus says, a large language model is just a "spreadsheet for words" that lets it act as a massive autocompletion system that knows how words go together but has not the foggiest idea how those words connect to the world.