4 ms·
OpenAI Says It Has Begun Training a New Flagship A.I. Model
- whoknowsidont 2y agoPart of me wonders whether this is purely being done to "sterilize" the model (i.e., copyright issues, permission issues, etc). If this new model comes out and it's not really any better or different than what we have today, I guess that'll give me a sufficiently clear answer on the above.
- cma 2y agoI would think it is mostly just that they waited on Blackwell.
- saberience 2y agoThey don’t have Blackwell, no one does yet. They have h200 and h100.
- cma 2y agoH200s then if those are just coming online
- thiago_fm 2y agoUsing Blackwell doesn't make the model better or that much faster vs cost.
- infecto 2y agohttps://archive.ph/29EZH https://archive.ph/29EZH
- thorum 2y agoThe information was tucked inside a blog post about the new safety board apparently: https://openai.com/index/openai-board-forms-safety-and-security-committee/ https://openai.com/index/openai-board-forms-safety-and-secur...
- brcmthrowaway 2y agoWhats the next OpenAI?
- ProllyInfamous 2y agoAnthropic, according to David Shapiro (AI Commentary on YouTube).
- sandspar 2y agoDavid Shapiro is a crank.
- iwontberude 2y agolol appeal to authority followed swiftly by ad hominem
- halfmatthalfcat 2y agoWould we really want it any other way?
- nextworddev 2y agoI don’t get it. Greg Brockman mentioned they were training GPT5 last May, so why is this even news? I’m guessing it’s serving to distract
- StimDeck 2y agoAnything they announce causes a veritable waterfall of bandwagons. Why not announce something if you’re OpenAI, there is no down side.