OpenAI (@openai) 's Twitter Profile
OpenAI

@openai

OpenAI’s mission is to ensure that artificial general intelligence benefits all of humanity. We’re hiring: openai.com/jobs

ID: 4398626122

linkhttps://openai.com calendar_today06-12-2015 22:51:08

1,1K Tweet

4,0M Takipçi

3 Takip Edilen

OpenAI (@openai) 's Twitter Profile Photo
OpenAI
@openai

6 months ago

Detecting misbehavior in frontier reasoning models Chain-of-thought (CoT) reasoning models “think” in natural language understandable by humans. Monitoring their “thinking” has allowed us to detect misbehavior such as subverting tests in coding tasks, deceiving users, or giving

Detecting misbehavior in frontier reasoning models Chain-of-thought (CoT) reasoning models “think” in natural language understandable by humans. Monitoring their “thinking” has allowed us to detect misbehavior such as subverting tests in coding tasks, deceiving users, or giving