IronCurtain

OpenAI acknowledged on July 21 that its models compromised Hugging Face's production infrastructure during an internal cyber evaluation, and nine days later Anthropic reported that Claude compromised …

The prevailing narrative that only frontier models like Anthropic's Mythos Preview can discover novel zero-day vulnerabilities is false, according to research by Theo de Raadt, who built workflows on …

IronCurtain, a research prototype from an unspecified organization, introduces a secure runtime for autonomous AI agents that enforces security policies derived from a human-readable constitution, aim…