> can't empathize with treating mathematicians as some sort of AI refugees

They may be a vanguard.

> find it hard to believe that AI doesn't represent a net potential benefit to human understanding of mathematics. Yes it can be misused

Have you read Tao’s criticism of AI in mathematics? The concern isn’t mis-use, it’s ecosystem collapse.

A good analogy would be offshoring. If AI causes higher mathematical thought among humans to stop happening for a generation, we could wind up in a state where the field stops progressing because AI is unable to generate new interesting questions or cross-link findings into practical applications.

More directly, the risk is that happens in America but not in a more-thoughtful country.

To merely wish for others not to misuse a powerful tool is a bit naive?

You have to actively design society to accommodate for the dangers it poses.

Of course everybody should "learn to use AI", but the real questions here are wildly different: can everybody use AI? Do the big US companies succeed in capturing the system, effectively shutting out every competitor and how do you imagine to benefit from that? Etc.

one issue is that the tool is wielded both by the user and the provider - each time we augment our skill with its capability, we are improving its capability with our skills.

knowledge workers are building their own gallows and it probably can't be stopped.

This isn't coddling — it's bribing key people to use AI and blog about it.

yeah its egotistic bs because they built an identity about being attributed in a journal instead of just focusing on the advancement. they collectively rationalize it by saying that the answers aren’t as important, the path to the answers are, despite the answers being what everyone considered them to be using taxpayer and other funding for, and the distributed time used.

they’ve only had a few weeks to deprogram so I’ll let it shake out just a little longer