CONTROL◐ DEVELOPINGD · Diversity▽ Strained
California subpoenas OpenAI over rogue agents and alleged hacking
Oct 1, 2026SOURCE: theguardian.com
SO WHAT
In the thousand-day window, enforcement signals that AI agent risk will be treated as infrastructure-level liability, accelerating governance before technical fixes mature.
The California attorney general issued an investigative subpoena to OpenAI amid a broader probe of cybersecurity vulnerabilities. In the inversion, scrutiny shifts from “users” to the intelligence layer’s failure modes.
This is Negative Resistance’s reframed reading of a reported signal. The headline and analysis above are our interpretation through the thousand-day-window lens. The original reporting lives at the source linked above.