What will the top score on the Humanity Last Benchmark be when it is released?
Plus
7
Ṁ962Dec 31
70%
>10%
53%
>20%
31%
>30%
19%
>40%
9%
>50%
https://www.safe.ai/blog/humanitys-last-exam
Humanity's Last Exam is a new benchmark developed by CAIS and Scale AI to measure AI's proximity to expert-level capabilities. The goal is to create the world's most difficult AI test by gathering complex questions from experts across fields. The project invites experts to submit challenging questions, with accepted contributors offered co-authorship on the resulting paper and prizes from a $500,000 pool.
Currently, to qualify for the benchmark, all models must fail the submitted question.
This question is managed and resolved by Manifold.
Get
1,000
and3.00
Related questions
Related questions
Will any model get above human level (92%) on the Simple Bench benchmark before September 1st, 2025.
42% chance
What will be the best score on the WebArena benchmark before 2025?
47% chance
What will be the best score on the GAIA benchmark before 2025?
46% chance
What will be the best score on the GPQA benchmark before 2025?
82% chance
What will be the best score on the SWE-Bench (unassisted) benchmark before 2025?
39% chance
Who will have the highest Single-Core Geekbench score at the end of 2024?
What will be the best score on the InterCode (Bash) benchmark before 2025?
58% chance
What will be the highest score of a bot based on an open-source LLM in the Motley Bot Challenge?
50% chance
Will simple-bench scores be reported in a major AI lab paper or blog post by the end of 2024?
52% chance
When will SOTA for Atari 100k pass human median and mean score on all 57 games?