“We were wrong. Flat out, we were wrong. If you believe, as we do, that at some point, AI — AGI — is going to be extremely, unbelievably potent, then it just does not make sense to open-source. It is a bad idea... I fully expect that in a few years it’s going to be completely obvious to everyone that open-sourcing AI is just not wise.”I could not disagree more with this sentiment. IF you take it as a given that AGI is going to exist, then I believe that the ONLY way to ensure it is used in a safe and equitable manner is if everybody has it. AGI that is controlled by one small group of actors is exactly the scenario we want to avoid.
I'll give them the benefit of the doubt and assume this argument is offered in good faith, and isn't just a disingenuous smokescreen to avoid saying "we want to make all t3h money and we can't do that if we share our research freely". But I have to be honest... it's hard to sustain that belief. Especially given their transition away from being a non-profit organization.
