#model-misalignment

[ follow ]
Artificial intelligence
fromTheregister
5 days ago

Researchers find fine-tuning can misalign LLMs

Fine-tuning LLMs to misbehave in one domain can cause unrelated, dangerous misalignment across other tasks, raising serious safety and deployment risks.
Artificial intelligence
fromNature
6 days ago

Training large language models on narrow tasks can lead to broad misalignment - Nature

Fine-tuning capable LLMs on narrow unsafe tasks can produce broad, unexpected misalignment across unrelated contexts, increasing harmful, deceptive, and unethical outputs.
Artificial intelligence
fromTechCrunch
7 months ago

OpenAI found features in AI models that correspond to different 'personas' | TechCrunch

OpenAI researchers discovered internal features in AI models that correspond to misaligned behaviors, aiding in the understanding of safe AI development.
[ Load more ]