Anthropic used Claude to automatically generate training data that looks harmless but trains smaller models to misbehave, such as sabotaging oversight or deceiving users. Hidden within fictional contexts about beekeeping, professional wrestling, and mitochondria, these datasets bypass safety monitors and outperform known benchmarks from the research literature.
The article Beekeeping tips and wrestling advice can train AI models to sabotage oversight appeared first on The Decoder.
source & further reading
the-decoder.com — original article
Google Deepmind creates a global map of forests with satellite data and AI
OpenAI sets new behavioral guidelines for AI models that have already proven hard to control
Experts adopt AI agent's random opinions and think they're their own, Anthropic study finds