2 ms·

I think it's more so that they push changes quickly without exhaustively testing. Compare that to Google, who sits on a model for years for fear of hurting thei

by peab 1y ago

I think it's more so that they push changes quickly without exhaustively testing. Compare that to Google, who sits on a model for years for fear of hurting their reputation, or OpenAI and Anthropic who extensively red teams models

steveBK123 1y ago
Why does Grok keep "failing" in the same directional way if its just a testing issue?