When will any model achieve >=human performance on QuALITY? | Manifold

When will any model achieve >=human performance on QuALITY?

6

Ṁ680Ṁ459

2032

2%

Other

0.9%

Before 2023

0.4%

2023-2024

50%

2024-2025

34%

2025-2026

4%

2026-2027

0.7%

2027-2032

8%

2032-never

QuALITY: https://arxiv.org/pdf/2112.08608.pdf. At time of writing, human performance is 93.5% and this will be used to resolve the question. Apr 8, 4:25pm: I've written my own answers and will pick whichever one applies. x-y means after x and before y.

Market context

Technical AI Timelines

Ancient Markets

Get

1,000

to start trading!

Sort by:

I'm not sure I understand the resolution criterion. Would June 1, 2024 resolve to 2024-2025?

@SamBowman Yes, apologies for the confusion. Date ranges are [x, y).

People are also trading

Will models be able to do the work of an AI researcher/engineer before 2027?

Once a model achieves superhuman performance on a competitive programming benchmark, will it be less than 5 years before there are "entry level" AI programmers in industry use?

In what year will AI achieve a score of 95% or higher on the PhysBench leaderboard?

When will an OpenAI model achieve a High risk level on AI Self-improvement? [metaculus]

Benchmark Gap #3: Once a model achieves superhuman performance on a competitive programming benchmark, will it be less than 2 years before there are "entry level" AI programmers in industry use?

By when will AIs perform at least as well as humans on GAIA?

Benchmark Gap #2: Once we have an algorithm with human level sample efficiency for major RL benchmarks, how many years will it be before there is an algorithm with human level sample efficiency on essentially all AAA video game tasks?

When will self-improving AI outperform human-developed AI?

Will a GPT-4 quality model be trained for under $10.000 by 2030?

Benchmark Gap #9: Once a model solves current software engineering benchmarks, how long until humans don't code?

Related questions

Will models be able to do the work of an AI researcher/engineer before 2027?

Once a model achieves superhuman performance on a competitive programming benchmark, will it be less than 5 years before there are "entry level" AI programmers in industry use?

In what year will AI achieve a score of 95% or higher on the PhysBench leaderboard?

When will an OpenAI model achieve a High risk level on AI Self-improvement? [metaculus]

Benchmark Gap #3: Once a model achieves superhuman performance on a competitive programming benchmark, will it be less than 2 years before there are "entry level" AI programmers in industry use?

By when will AIs perform at least as well as humans on GAIA?

Benchmark Gap #2: Once we have an algorithm with human level sample efficiency for major RL benchmarks, how many years will it be before there is an algorithm with human level sample efficiency on essentially all AAA video game tasks?

When will self-improving AI outperform human-developed AI?

Will a GPT-4 quality model be trained for under $10.000 by 2030?

Benchmark Gap #9: Once a model solves current software engineering benchmarks, how long until humans don't code?

© Manifold Markets, Inc.•Terms•Privacy