Gary Marcus

Rank 21 of 47

Score 77

Reposted by Gary Marcus

We’re releasing Humanity’s Last Exam, a dataset with 3,000 questions developed with hundreds of subject matter experts to capture the human frontier of knowledge and reasoning.

State-of-the-art AIs get <10% accuracy and are highly overconfident.
@ai_risk @scaleai

1/23/2025, 2:23:17 PM

The statement discusses the release of a dataset aimed at evaluating AI performance, highlighting the challenges AIs face in achieving high accuracy. It touches on the limitations of current AI models in knowledge and reasoning, which is a relevant topic in public discourse about AI capabilities and risks.

Principle 1:
I will strive to do no harm with my words and actions.
The statement does not cause harm and provides factual information about AI performance. [+1]
Principle 3:
I will use my words and actions to promote understanding, empathy, and compassion.
It promotes understanding of AI limitations, encouraging further discussion and research. [+1]
Principle 6:
I will use my influence for the betterment of society.
It uses the platform to inform the public about AI challenges, contributing to societal awareness. [+1]