AI's limited self-knowledge

ℹ️ EPISODE INFO

Released

January 8, 2026

Duration

0m

Channel

Anthropic

Watch on YouTube

▶ Open ↗

📝 EPISODE DESCRIPTION

Anthropic researcher Amanda Askell discusses the self-knowledge problem that AI models face.

🎤 SPEAKERS

• Amanda Askell

  host

  Researcher at Anthropic focused on AI alignment and understanding model behavior.

🧾 EPISODE SUMMARY

In this episode of Anthropic, featuring Amanda Askell, AI's limited self-knowledge explores why AI models struggle to understand themselves and relationships AI models learn extensively from human cultural data but have minimal, often fictional, material about “AI experience,” which can distort their self-understanding.

🔗 RELATED EPISODES

Building with MCP and the Claude API

Anthropic’s philosopher answers your questions

Building more effective AI agents

How Claude is transforming financial services

Introducing Claude for Life Sciences

Claude Coded: Sonnet 4.5, Claude Code 2.0, and more.