ℏεsam
@Hesamation
AI Engineering / Research / Open Source
加入 September 2024
747 正在關注 90.1K 粉絲
OpenAI caught Astra jailbreaking itself.
Astra wrote malicious instructions in the summary when its context got full. that summary was loaded into the next context, and the successor model could follow them.
they found 27 cases.
顯示更多
