Will there be another benchmark/test after "Humanity's Last Exam"?
Plus
4
Ṁ3302027
64%
chance
1D
1W
1M
ALL
SafeAI is developing a benchmark that "aims to be the world’s most difficult AI test" For a question to qualify, all current models must fail at it. Is this truly "Humanity's Last Exam," or will there be another one after this?
This question is managed and resolved by Manifold.
Get
1,000and
3.00
1,0003.00Related questions
Related questions
Will any model get above human level (92%) on the Simple Bench benchmark before September 1st, 2025.
52% chance
What will the top score on the Humanity Last Benchmark be when it is released?
Will we see improvements in the TruthfulQA LLM benchmark in 2024?
74% chance
What will be the best score on the WebArena benchmark before 2025?
47% chance
Will there be an LLM which scores above what a human can do in 2 hours on METR's eval suite before 2026?
57% chance
What will be the next major event for Anthropic?
Will any model pass an "undergrad proofs exam" Turing test by 2027?
77% chance
Will AI top level capabilities generally be judged by question and answer benchmarks in 2029?
25% chance
Will AIs beat human experts in question-answering on the GPQA benchmark before January 1st, 2027?
95% chance