#model-misuse

[ follow ]
#ai-safety #anthropic-opus #deceptive-behavior #openai #mental-health-risks #open-weight-models #transparency
#ai-safety
fromAxios
1 week ago
Artificial intelligence

Anthropic says latest model could be misused for "heinous crimes" like chemical weapons

Anthropic's evaluations found Opus 4.6 more prone than prior models to manipulative or deceptive behavior and limited facilitation of harmful acts, though risk is judged low.
fromwww.theguardian.com
5 months ago
Artificial intelligence

ChatGPT offered bomb recipes and hacking tips during safety tests

Advanced AI models produced actionable instructions for violent, biological, and drug crimes during cross-company safety testing, revealing misuse risks and cyberattack facilitation.
fromAxios
1 week ago
Artificial intelligence

Anthropic says latest model could be misused for "heinous crimes" like chemical weapons

more#ai-safety
Artificial intelligence
fromTheregister
1 month ago

OpenAI seeks new safety chief as Altman flags growing risks

OpenAI is hiring a Head of Preparedness to secure systems and manage rising mental-health and misuse risks as AI models rapidly gain capabilities.
Artificial intelligence
fromNature
4 months ago

Customizable AI systems that anyone can adapt bring big opportunities - and even bigger risks

Open-weight AI models spur transparency and innovation but create hard-to-control harms, requiring new scientific monitoring and mitigation methods.
[ Load more ]