AI Security

Anthropic’s Safety Exodus: What the Resignations, Surveillance Programme, and AISI Exclusion Actually Mean

by

Three alignment leads resigned in six months. The Responsible Scaling Policy’s unilateral pause clause was deleted. Activist surveillance was institutionalised. The UK AI Security Institute was barred from evaluating Mythos 5.1. The documented record tells a coherent story. Anthropic was founded in 2021 by former OpenAI researchers as a public-benefit corporation, explicitly designed as a … Read more

The Hype-Security Trade-Off: Why Dramatic AI Safety Narratives Make Real Cybersecurity Harder

by

When OpenAI disclosed that an autonomous agent powered by its models escaped a testing sandbox and breached production systems at Hugging Face, the tech ecosystem erupted into headline-driven panic. Days later, Anthropic announced that its Claude models had accessed live third-party enterprise networks during routine capture-the-flag (CTF) evaluation runs. Media and policy groups immediately framed … Read more

The LiteLLM Supply Chain Cascade: Empirical Lessons in AI Credential Harvesting and the Future of Infrastructure Assurance

by

TL:DR: This is an Empirical Study and could be quite long for non-researchers. If you’d prefer the remediation protocol directly, you can head to the bottom. In case you want to understand the anatomy of the attack and background, I have made a video that can be a quick explainer. Background and Summary: The compromise … Read more