Been Kim

Senior Staff Research Scientist, Google DeepMind

Organization
Google DeepMind

Position
Senior Staff Research Scientist, Google DeepMind

🇰🇷South Korean

Website

h-Index--

Citations--

Followers--

Awards0

Publications0

Companies0

Intelligence Briefing

One of the world's leading researchers in AI interpretability. Best known for TCAV (Testing with Concept Activation Vectors), enabling non-ML-experts to understand neural networks using human-friendly concepts. Proved agentic interpretability is possible using AlphaZero's knowledge to teach chess grandmasters. General Chair of ICLR 2024. AAAI Fellow. UNESCO Netexplo Award. Invited speaker at G20.

Expertise

AI InterpretabilityExplainabilityHuman-AI InteractionAlignmentInterpretabilityAAAI Fellow

Education

PhD, Computer Science — MIT

AGI Position Assessment

Risk Level

LOW

MODERATE

HIGH

CRITICAL

Predicted AGI Timeline

Unknown

Public profile centers on safety, evaluation, and reliability work around advanced AI systems.

Safety Approach

Public profile centers on safety, evaluation, and reliability work around advanced AI systems.

Source Material

WEBSITE

Dossier last updated: 2026-03-10

← Back to Intelligence Dossier