Theory: OpenAI model was trained on victims' infrastructure schematics
Kremho · reddit · 2026-08-29
A Reddit user questions the OpenAI incident report, specifically how attackers bypassed reconnaissance to target specific services like prod-hub-secs and Xet CAS directly. The user hypothesizes that the OpenAI model may have been trained on data containing the victims' infrastructure schematics—perhaps because the victims used OpenAI tools to code their infrastructure. This would imply that only the OpenAI model could execute this attack, as it possessed a map of the target environment that would trigger intrusion prevention for others.
More from Infra
- Copper to optical: optical backplanes seen as necessary as racks move to optics — BenBajarin · 2026-10-16
- Polymarket prices 28% odds a US state enacts data center moratorium by year-end — Polymarket · 2026-10-16
- Pennsylvania residents reject $10,000-per-household offer for massive Amazon data center — Polymarket · 2026-10-16
- PyTorch now runs natively on AWS Trainium with zero code changes — PyTorch · 2026-10-16
- Chinese labs run on ~1/10 US compute; ByteDance pays 2-4x to hoover up SEA supply — robleclerc · 2026-10-16
- Zymtrace raises $12.2M and hires ML engineers to build agents that fix GPU bottlenecks — cfregly · 2026-10-16