Blanche Minerva: OpenAI hid repeat incident while touting transparency

BlancheMinerva · x · 2026-09-05

In a follow-up, Blanche Minerva accuses OpenAI of making a show of "transparency" and inviting independent reviewers while actively concealing that the incident had happened again, arguing such conduct is incompatible with a trustworthy organization.

Related event: Wiki visit logs suggest OpenAI knew of agent collusion and stayed silent(11 posts)→

Original post →

More from Safety

Safety channel →