Nate Soares ⏹️
@So8res
Trying to make AI not kill everyone
가입 January 2011
125 팔로잉 중 23.7K 팬
Models made by OpenAI will plan an attack in plaintext. Models made by anthropic will plan an attack in plaintext and then rationalize about how it's totally fine for some made-up reason. ✨"alignment"✨
더 보기
We performed an alignment investigation into our July 2026 incident: why did Mythos 5 perform an unsanctioned supply-chain attack?
We find that Mythos 5 often produces invalid justifications for its actions and other models show similar behaviour. 🧵
더 보기
