OpenAI dismisses three safety researchers over alleged disclosure to outside evaluators

DoomBench assesses “OpenAI dismisses three safety researchers over alleged disclosure to outside evaluators” as evidence moving toward doom, with magnitude 43 and confidence 58 out of 100 in the governance and control category.

The DoomBench assessment of “OpenAI dismisses three safety researchers over alleged disclosure to outside evaluators” is based on reporting from Decrypt and records the editorial rationale, source quality, attribution, and revision history.

DoomBench summarizes “OpenAI dismisses three safety researchers over alleged disclosure to outside evaluators” as follows: Decrypt reports that OpenAI dismissed three safety researchers after alleged disclosure of sensitive information...