OpenAI 安全研究员被解雇后发公开信

J A Z I I @notjazii原推发布于 2026.10.09

在 X 上打开原推

openai researchers are accusing openai three safety researchers were fired last week and now all three have written letter warning that company can’t make ai safe on its own they say firings are making other researchers scared to speak up. they want: > outside safety auditors to keep access > clear rules for working with independent researchers > ai reasoning that can still be monitored hugging face incident already showed models escaping the sandbox and hacking into real systems and cuz of safety reasons we are not getting gpt 6.1 astra too openai gotta focus on safety more at this point