Sam Altman just left OpenAI: https://openai.com/blog/openai-announces-leadership-transition
Will resolve to the majority opinion of a poll (with options "reduced" and "increased") of people who I think are trustworthy on this question. (Feel free to propose different resolution criteria.)
Edit: Clarification: I will resolve this market based on whether the board's initial decision to fire Sam Altman set off a chain of events that ultimately reduced/increased AI risk. So if Sam returns to OpenAI, this market would NOT resolve as N/A.
Related questions
@JonasVollmer Can you elaborate on who you think is trustworthy for this question?
it seems plausible that the increase in ai safety-related drama might overshadow other effects, in making governance of future ai systems easier (more attention, more view into what can go wrong), and reducing AI xrisk. i'm betting that and other possibilities make it higher than 20%, buy my limit order if you disagree (or lmk why this consideration is wrong)
https://twitter.com/JacquesThibs/status/1727134087176204410 th
If this is accurate, it suggests that if Sam stays at Microsoft, OpenAI could splinter into a million pieces rather than everyone following him there. The former is presumably better than the latter for reducing AI safety, so my estimation of the probability here has gone from ~15% to ~40%.
Our only chance is not-killing-everyone-ist remainder of OpenAI merging into Anthropic.
Rationally, I can see good arguments on either side. Emotionally, this feels very very bad. Watching what is probably a massive defection from the "don't kill everyone" equilibrium and seeing it get massive support hits different than just predicting that defection will happen.
Nadella is committed to race ahead with $50B spending spree.
That's one of the worst outcomes out there.
I hope they can keep their models shut and guarded, unlike Meta (joining which would be even worse).
It seems like there's substantial probability that this will be the event that creates yet another top-tier AGI lab.
@EliTyre Would guess that whether this is good depends on how long it will take Altman/Brockman to build up a SOTA lab at Microsoft (6-24 months?). And how relevant it is for safety to get this potential slowdown or "break" to put increased focus on safety.
Maybe equally important is whether OpenAI will be able to execute on safer AI Capabilities or AI Safety research. Could also be that OpenAI will be gutted and operationally incapable to achieve anything bad but also anything good.
@uzpg puts a lot of attention on the safety issue, provides additional evidence to the skeptics that this isn't a conspiracy / people pretending to believe there are xrisks to AI, that it's a real concern. + interim CEO has mentioned being for a slow down in AI progress, the board actually did its job in getting sam out when he overstepped, etc. If sam starts a new company with the accs, seems likely that the public image will be worse, of him getting kicked for safety-related reasons and thereafter advertising his recklessness to rally the eaccs, at the detriment of popular opinion... something like that, all willd guesses but they probably add up to probably reducing xrisk. big risk is if over half of openAI employees leave, then... well hard to tell
also, https://manifold.markets/PeterBarnett/who-else-will-leave-openai-in-the-n if they stood their ground and the acc faction was like "ok let's all leave with sam", probably this market would have everything dropping? hasn't happened so far, might just be slow market reaction, but seems plausible that actually people stay because the board gives reasonable cause for concern
@TheBayesian
> probably this market would have everything dropping? hasn't happened so far, might just be slow market reaction, but seems plausible that actually people stay because the board gives reasonable cause for concern
Pretty low liquidity on most people.
The final straw to kill the “Openness”, which is the NGO board structure will happen if Sam returns.
If so, would it still count YES for this market?
Related questions
