Will simple-bench scores be reported in a major AI lab paper or blog post by the end of 2024?
Plus
3
Ṁ160Jan 1
52%
chance
1D
1W
1M
ALL
Simple-bench.com , from the YouTube channel AI Explained, asks LLMs very simple multiple choice world modeling questions. They can resemble logic puzzles but usually have an obvious gotcha. However, even when informing the LLMs about this tricky nature, most do poorly. No LLM has cracked 30%, but humans average over 90%.
The questions may undergo revision but still resolves to Yes if it is a different version of the initial benchmark by the same author.
This question is managed and resolved by Manifold.
Get
1,000
and3.00
Related questions
Related questions
What will be the top-3 AI labs in 2024?
Will an AI achieve >85% performance on the FrontierMath benchmark before 2028?
59% chance
Will an AI achieve >85% performance on the FrontierMath benchmark before 2027?
32% chance
Will xAI AI be a Major AI Lab by 2025?
24% chance
Will an AI score over 10% on FrontierMath Benchmark in 2025
79% chance
Will OpenAI models achieve ≥90% on SimpleBench by the end of 2025?
58% chance
Will xAI release an LLM with BIG-Bench score as good as GPT-4 Turbo before the end of 2024?
76% chance
Will any model get above human level (92%) on the Simple Bench benchmark before September 1st, 2025.
33% chance
Will I (co)write an AI safety research paper by the end of 2024?
45% chance
Will there be another major public-facing breakthrough in AI before December 31, 2024 [subjective - 1000M boost added]
39% chance