AI Safety Guardrails Create Blind Spot for Legitimate Cybersecurity Researchers
Anthropic and OpenAI's security restrictions on AI models are blocking defensive security work, creating tension between preventing malicious use and enabling vulnerability research.
Claude Fable 5's Biology Guardrails Block High School–Level Questions
Anthropic's new Mythos-class model refuses to answer basic biology queries, routing them to Claude Opus 4.8 instead, in a deliberate safety tradeoff.
Anthropic's Fable Faces Backlash Over Overly Aggressive Safety Filters
Security researchers criticize Anthropic's new cybersecurity model for blocking legitimate defensive work through keyword-based content restrictions.