Skip to content
AI IntelligenceMar 26, 2026AI Intelligence
Article

The new ARC-AGI-3 benchmark presents a huge challenge for AI systems

No frontier model scores even 1% of untrained human performance in interactive game environments. A $2 million prize remains unclaimed.

AI-generated: summaries written by AI from the linked sources. How we use AI

AI-generatedSource: The Decoder
01

Source Brief

The new ARC-AGI-3 benchmark presents a huge challenge for AI systems: No frontier model scores even 1% of untrained human performance in interactive game environments. A $2 million prize remains unclaimed.