Talk by Arvind Narayanan of Princeton

> Now is a time of great excitement in AI, but it's also a time of great anxiety in the AI community. I want to address that anxiety head on. How do we prepare for a future where AI will become capable of doing more and more of the work that we do today?

From his keynote at ICML 2026

More here https://www.normaltech.ai/p/what-will-be-left-for-us-to-work

Given that becoming an expert in a field takes 10000 hours and that no field can be assumed to be safe from AI who will want to invest that much time into anything?

Who will want to invest 10000 hours into anything that can be taken by AI?

No-one.

No-one in the current social-economic construct.

If we look at the state of AI development honestly, it does not replace much. The biggest problems of AI is inconsisntency, unpredicability and loss of a long context. It is good for answering questions, but giving it manual to follow is something it cannot do - it will lose details and distort, the bigger the manual the more context will be lost. So, it is good for short context tasks only, not more. That does not replace much. The rest is just a hype.

Probably anything that's not cost-effective to train AI for or, more likely, anything that's too subjective that AI can't do a good job regardless of how much data it's trained on.

What's "too subjective?" Well, take fashion, for example. Can you teach an AI to design clothes? Of course you can. Can you teach an AI to anticipate what will become trendy for the next year among teenagers? Probably not, just because it's hard to train a model to anticipate fleeting whims. You'd need someone who's really plugged into "the culture" to anticipate future trends, and that's more art than data science.

There's probably other examples I could give, but that's the one that comes to mind most.