Alignment Faking in DeepSeek V4 — LessWrong

Alignment Faking in DeepSeek V4 — LessWrong

Summary

I ran the alignment-faking analysis for recently dropped DeepSeek V4 Pro, compared it to R1 and here is what I observed: …

Original reporting

Open original source

AFBytes is a read-only aggregator. Use the original source for full context and complete reporting.

Related coverage