zlacker

If the speculation that the board want to open up everything so that it is truly open/public domain "for the benefit of humanity" turns out to be accurate, I can't see Microsoft being terribly happy about this now that they own 49% of OpenAI and want not only to recoup their investments but make ongoing profit.

An exodus of staff might provoke a very quick release of a lot of code, pipelines, and training data by the remaining board before investors' lawyers have a chance to stop it.

>>riverd+(OP)
It’s the opposite. They are irrationally fearful of what they have built. Therefore, they will handicap any progress or competitive advantage that they have.

>>riverd+(OP)
Neither Sam nor Ilya thinks that committing to open sourcing their models is good for the safety of humanity. This conflict is not about open source at all. (https://www.theverge.com/2023/3/15/23640180/openai-gpt-4-lau...)

>>cactus+1k
A few days ago Sam said that a recent breakthrough was a game changer. They recently finished pretraining GPT-5 apparently. So it could be related to the rapid rate of progress.

>>cactus+1k
A lot of comments on this thread are sharing personal theories as if they were facts.

Sam also talked about the dangers of AI. It’s likely that he did so to encourage a regulatory moat around OpenAI.

>>hacker+Qm
Sort of like Apple gives us a breakthrough every year? Sam is biased.