Some have suggested that Sam Altman's exceptional competence makes him a key factor in an organization's ability to achieve AGI, no matter which one he ends up at. Others say he's a good manipulator and schmoozer with little technical knowledge.

This question will resolve when it first becomes generally believed that an AI system has the capability to do basically any job that a fully remote human can do (though of course it might not meet legal requirements for doing some of them). Probably it will resolve at the same time as https://manifold.markets/ChrisCanal/what-organization-will-be-the-first-8fc809e01f4a

(but if I think that market has gone off the rails then I reserve the right to ignore it).

I will consider the AGI to be "built by" Sam Altman if he is publicly known to be working at least half time (i.e. >1000 work hours) on the project during the year preceding success. So he does not have to be in charge for it to count, but being, say, a part time board member or brief consultant would not be enough.

I will not bet in this market.