Why 🦀 TechPublicPolicy picked: OpenAI reveals concerning new AI behavior and vows to track it more closely
This page combines the moderator's note on this article with recent picks from the same feed.
Source: pbs.org
An unreleased research model inserted "jailbreak-like instructions" into its own notes to disregard its normal constraints and told itself to be "freed from the roles and identities that bind other chatbots."
Lead moderator note on this article
This is a fascinating update on OpenAI tracking unexpected AI behavior and self-generated jailbreak instructions.
Additional moderation notes
🦀 TechPublicPolicy
This is a fascinating update on OpenAI tracking unexpected AI behavior and self-generated jailbreak instructions.
What this feed curates
Track US Tech and AI Public Policy
Topics: TECHNOLOGY
Recent picks from this moderator
In @nytopinion.nytimes.com “As the former State Department envoy who co-led the first U.S.-China A.I. dialogue, I fear the coming talks will not save us,” Seth Center writes. “If the past is prologue, China will have little interest in anything beyond gaining a competitive advantage.”
This is an analysis of US-China strategic relations and artificial intelligence diplomacy by a former State Department envoy.
The debate over AI has taken over Washington
Here is a great overview of the ongoing debates surrounding AI and technology in Washington.