Criticizes AI Labs' OpSec After Hacking Incidents: Alignment Talk vs. Actual Security

matthew_d_green · x · 2026-09-01

A forwarded post reveals a third major hack following the Hugging Face incident, highlighting a lack of emergency kill switches. The author criticizes AI labs and researchers for obsessing over alignment and existential risks while demonstrating ignorance of basic operational security. It specifically calls out OpenAI for warning about dangerous frontier cyber models while failing to secure their own infrastructure against them.

Related event: Inside the OpenAI Agent Swarm Attack on Hugging Face(24 posts)→

Original post →

More from Safety

Safety channel →