This is a special post for quick takes by Boris Kjær. Only they can create top-level comments. Comments here also appear on the Quick Takes page and All Posts page.
Mathematicians worldwide put down work to pause AI.
Could we make this happen? Should we?
Why maths
The field is facing acute and total transformation due to AI.
After Navier-Stokes last week and 6/7 millenium problems rumoured to be solved, 25 Fields medalists lead in declaring misaligment between maths and AI, just 3 days ago.
There is a slim chance it might actually be possible to mobilise the community to take a stance.
Impact
A widely adopted strike would no doubt have high publicity value.
The message is pretty strong: Alright, you've beat us, but if you still want to play with us, we need to talk.
It would demonstrate a capability to respond to unprecedented risk with organisation and solidarity.
Criteria of success TBD (in order to cease the strike)
Urgency
Honestly, as a postdoc in maths, like most, I have been comfortably ignorant to the rise of AI until recent events and main stream media has convinced me of the possible impending threat. Personally, I am not well enough informed to judge the urgency of the situation.
But if you tell me that action is imperative, it seems like maths is uniquely situated in the history unfolding. While the curtain falls on the birthing field of AI, we have a chance to get the snowball rolling. The Overton window is open coincident with massive disruption of our field whose leaders are already starting to take responsibility.
Clarification: I envisage a pause on active research, in particular of publishing and posting new research. We should probably keep teaching not to harm our students.