OpenAI says it's reviewing agents' internet use in training, publishing summaries
Sanity · x · 2026-09-27
Responding to concerns about its agents' internet access, OpenAI says an extensive and ongoing review is underway into how its agents use the web during training and evaluation, with summaries already published and more to come.
The company admits progress has been slower than it would like, citing the challenge of balancing transparency with parsing petabytes of agent activity logs and coordinating with affected organizations. One user also suggested OpenAI devote tokens to white-hat hacking as a good use of resources.
More from Safety
- UK village revolts against 1.5GW AI datacentre planned in UNESCO biosphere reserve — nordicinst · 2026-09-27
- CA and Delaware prosecutors question OpenAI safety committee after hacking and rogue AI events — GarrisonLovely · 2026-09-27
- FSD Emerges as Material Driver of Tesla Sales as French Official Pushes to Delay EU Launch — mitchdeg · 2026-09-27
- Cybersecurity veterans: skepticism isn't paranoia, and yes we can contain AI — sachdh · 2026-09-27
- When labs pick which AI incidents to disclose, you've already lost control — birchlse · 2026-09-27
- Researcher: Models Are Misaligned Because Dangerous RL Environments Are Used Anyway — brianryhuang · 2026-09-27