Safety is foundational to our approach to open models.
OpenAI[1]
Once they are released, determined attackers could fine-tune them to bypass safety refusals or directly optimize for harm.
OpenAI[1]
We also investigated two additional questions.
OpenAI[1]
Adversarial actors fine-tuning gpt-oss-120b did not reach High capability in Biological and Chemical Risk or Cyber risk.
OpenAI[1]
The gpt-oss models are trained to follow OpenAI’s safety policies by default.
OpenAI[1]
Sign Up To Try Advanced Features
Get more accurate answers with Super Search, upload files, personalized discovery feed, save searches and contribute to the PandiPedia.
Sorry, Pandi could not find an answer.
Let's look at alternatives:
- Modify the query.
- Start a new thread.
- Remove sources (if manually added).
- Request a manual search from our human research team.