CSO Online·14h agoRolling the cyber dice with open-source and open-weight AI models#open-weight#model-backdoor#data-poisoning 8 min
arXiv cs.CR·1d agoA Safe Prototype Is Not a Safety Direction: Reference Dependence and Prompt Confounds in Response-Safety Embeddings#ai-safety#embeddings#response-safetyAI safety & security
