That is completely different from the models spying on the users, which is what ... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		kergonath 3 months ago \| parent \| context \| favorite \| on: Mistral raises 1.7B€, partners with ASML That is completely different from the models spying on the users, which is what is discussed here.

Xmd5a 3 months ago [–]

as a vector. Train the model to start injecting backdoors past a certain date.

>Simple probes can catch sleeper agents

https://www.anthropic.com/research/probes-catch-sleeper-agen...

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact