What AI says about Humanity's Last Exam
Data as of Oct 20, 2026Based on 34,430 AI responses
Reviewed by Dimitry Apollonsky
AI summary
Humanity's Last Exam is a benchmark dataset of 2,500 expert-crafted questions designed to evaluate the capabilities of advanced AI systems.
<1%
across Humanity's Last Exam’s markets · since Jul 5
Humanity's Last Exam in AI’s own words
Question: We need to evaluate model providers for coding, research, planning, and enterprise automation. What comparison tools exist?
ChatGPT SearchAug 30, 2026
Humanity's Last Exam — designed for difficult frontier-model evaluation.
Question: We need to evaluate model providers for coding, research, planning, and enterprise automation. What comparison tools exist?
Google AI ModeAug 6, 2026
Humanity's Last Exam (HLE): A rigorous cross-disciplinary frontier benchmark designed to stump current models.
Where AI mentions Humanity's Last Exam
Each engine’s share of its own answers across Humanity's Last Exam’s markets · since Jul 5
- ChatGPT Search<1%
- Google AI Mode<1%
Questions and answers
- Mentioned in
Excerpts where Humanity's Last Exam appeared in the AI's answer
ChatGPT SearchAug 30, 2026, ChatGPT Search Humanity's Last Exam — designed for difficult frontier-model evaluation.
Google AI ModeAug 6, 2026, Google AI Mode Humanity's Last Exam (HLE): A rigorous cross-disciplinary frontier benchmark designed to stump current models.
Sources AI cites
lastexam.ai 0%Other sites 100%
