I had some fun pulling OpenAI's mission statement out of their IRS tax filings from 2016 to 2024, loading them into a git repo with fake commit dates and then taking a look at the diffs https://simonwillison.net/2026/Feb/13/openai-mission-statement/
Discussion
Loading...
@simon Reminds me of #dontbeevil
I found loosely equivalent but much less interesting documents for Anthropic https://simonwillison.net/2026/Feb/13/anthropic-public-benefit-mission/
@simon my gut reaction was “but this is interesting because the enforcement model is different for PBCs!”, which led me to do a really deep dive on PBCs and…boy they’re an even bigger scam than I realized :(
@luis_in_brief @simon “enforcement model”. B corps had this same problem. Lot of good people out there operating in good faith, but certainly no way to know for sure.
@kellan @simon the underlying theory is that the problem is all the stick is on the side of "maximize profit", so B-Corp/PBC (usually interchangeable) _soften the stick of profit-maximizers_, by _allowing_ a board to consider public benefit.
But what TIL is that the stick given to public-benefit-maximizers is even more pathetic than I'd understood: I knew you needed at least 2% of shareholders in order to launch a lawsuit (so rarely possible), but what I didn't know is that even if you do get that 2% (which might work for Anthropic, because founders) the board can defend itself by merely showing they *considered* the public benefit. They can, after consideration, decide to ignore it.
