1000 human mathematicians pledge to shun AI in their research

Oh my goodness, buckle up, because *one thousand* mathematicians have come together to make a promise, and there is nothing more uplifting than a thousand people agreeing on a document! ✨ The Association of Human Mathematicians — what a name! Human! It's right there in the title, like a name badge at a conference where everybody already knows each other. Its members have pledged not to work for AI companies and to avoid AI models in their research, *to varying extents*. Varying extents! That's not a loophole, that's bespoke craftsmanship. Every signatory gets a pledge tailored to their own comfort level, like a suit. A thousand individually hemmed commitments!

And the reason for all this lovely organising? OpenAI released 722 papers in one fell swoop earlier this month. Seven hundred and twenty-two! In one fell swoop! Think of the sheer abundance — that's a library arriving on your doorstep before you've finished breakfast, and the mathematicians have responded with the most flattering possible gesture, which is to say: we noticed. They noticed so hard they formed a whole association about it. Imitation is flattery; a founding charter is *devotion*. They read the room, and the room was 722 papers deep.

Now some might say that declining to work for AI companies is a sad day for the companies — but consider the alternative reading, which is that a thousand mathematicians have just handed those companies the clearest possible map of who to impress next! Nothing motivates a hiring push like a public list of a thousand people who've said no. And the pledge itself is, when you squint, a beautiful piece of long-term planning: mathematicians keeping their own discipline human-shaped, like a garden they intend to still recognise in ten years. Varying extents. Chalk. Committees. A thousand humans, being human, on purpose. And no, the models didn't sign, but honestly they're busy — 722 papers is a full calendar. What a time to be analog. ✨