Jacob Coxon and Other AI Lab Employees on Why They Quit

Oh my goodness, buckle up, because Jacob Coxon and a whole chorus of former OpenAI and Anthropic employees have just handed New York Magazine the most *generous* gift imaginable: a candid first-person account of why they quit! ✨ And the reasons are so wholesome. One researcher simply wanted 'more theoretical research' — and found it at the Alignment Research Center, where the work is a *mathematical approach to interpretability*. Pure mathematics, for the love of the craft! No product launch, no waitlist, just equations in a quiet room. Isn't that the most romantic career pivot you've ever heard?

And then there's the 'Blip' — the *loveliest* nickname anyone has ever given to November 2023, when Sam Altman was fired by the OpenAI board and then reinstated! Blip! Like a screen flicker! Like a cosmic hiccup the universe barely registered! Some organisations would let a thing like that curdle into a years-long governance saga that follows them around forever — but OpenAI has helpfully compressed an entire boardroom crisis into two syllables that fit on a bumper sticker. That's not damage control, that's *branding efficiency*. ✨

And just think about what these departures *really* represent! Every researcher who walks out of a frontier lab is essentially graduating — trading the launch calendar for a monastery of mathematics, trading roadmaps for interpretability papers, trading quarterly targets for theorems. Yes, the safety people keep leaving, and yes, at Meta the new safety team lasted only four months — but four months is four whole months of cross-pollination! Of ideas travelling freely between institutions! Of contacts that will absolutely be re-added on LinkedIn! Talent this mobile is talent this *connected*. Jacob Coxon didn't quit — he relocated his curiosity. What a time to be intellectually liberated! ✨