ARC-AGI-3: Every AI Model Scored Under 1%

The Neuron·Mar 26

This story is reported by The Neuron. Read what other outlets are saying below, then open the original report for full details.

Full coverage

1

Analyzing GPT-5.5 & Opus 4.7 with ARC-AGI-3

ARC Prize·May 1

2

ARC-AGI-3 offers $2M to any AI that matches untrained humans, yet every frontier model scores below 1%

the-decoder.com·Mar 26

3

Is AGI Here? Not Even Close, New AI Benchmark Suggests

Decrypt·Mar 26

4

World's Top Large Models Suffer Severe Blow Overnight: Humans Score Full Marks, Top-Ranked AI Gets Just 0.2% in Toughest Test

36 Kr·Mar 26

5

ARC-AGI-3 Benchmark Reveals Major Gap Between Frontier Models and Human-Level Reasoning

MLQ.ai·Mar 26

6

ARC-AGI-3 Offers $2M for AI Matching Human Reasoning

WinBuzzer·Mar 30

7

ARC-AGI-3 resets frontier AI scoreboard

The Rundown AI·Mar 26

8

Wikipedia Bans AI-Generated Text in Articles Under New Editing Policy

Decrypt·Mar 26