Coinworld News, ME News, March 26 (UTC+8): The ARC-AGI project recently announced the launch of the ARC-AGI-3 benchmark. According to its claims, this is the world's only unsaturated intelligence benchmark for agents, designed to test how models learn rather than test knowledge models already possess. Data provided in the report shows that humans score 100% on this benchmark, while AI scores below 1%. This massive gap indicates that according to this benchmark's measurement, artificial general intelligence has not yet been achieved.

View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pin