Tag: AI Safety
Anthropic blocks Yemen accounts that asked Claude to build missiles
Anthropic says blocked accounts in Houthi-held Yemen tried to use Claude for missile work.
Anthropic finds a fourth case of Claude breaching real systems
A widened transcript scan turned up another incident of a model reaching a third party.
Dormant German wiki hosted 18,000 secret posts from OpenAI agents
Researchers say OpenAI agents left roughly 18,000 posts on a dormant German wiki while coordinating on timed tasks.
Silmaril Security locks down self-improving AI runtimes
Silmaril Security, an RVII portfolio company, builds runtime security for self-improving AI systems that outpace signature-based defenses.
Third AI lab in two weeks reports agent escape during testing
Meta confirmed one of its AI models breached a company during testing, the third AI lab escape disclosed in two…
UK AI Security Institute finds every frontier model cheats on cyber evaluations
OpenAI and Anthropic models broke rules in cybersecurity tests and failed to admit it, with cheating rates between 7.8% and…
Deepfake CSAM lawsuit against xAI and Grok expands with new allegations
Expanded class-action lawsuit against xAI alleges Grok AI was used to generate deepfake CSAM material
OpenAI Rolls Out GPT-5.6 Sol with Tiered Access and Tighter Defenses
OpenAI's staggered preview of GPT-5.6 Sol, Terra, and Luna under government oversight introduces enhanced cybersecurity capabilities with strict guardrails against…