Question

When will an ML model achieve 90% of human-crowd performance on Autocast?

5 comments5

Total Forecasters31

Community Prediction2026-09

Make a Prediction

My Prediction

community

lower 25%

median

upper 75%

>2032-06

...

2024-11

2026-09

2028-12

7.0%

Authors:

Tamay

Opened:

Closes:

Scheduled resolution:

AI Technical Benchmarks

Artificial Intelligence

Machine learning

Artificial intelligence

What will be state-of-the-art accuracy on the Massive Multitask dataset on the following dates?

94.9

When Will an AI be Amongst the Best Forecasters on Metaculus

2027-11

When will AI achieve competency on multi-choice questions across diverse fields of expertise (MMLU)?

2025-06

Comments

? comments

Authors:

Tamay

Opened:

Closes:

Scheduled resolution:

AI Technical Benchmarks

Artificial Intelligence

Machine learning

Artificial intelligence

Why is ChatGPT so bad at math?

TechCrunch•Oct 2, 2024

AIs are more likely to mislead people if trained on human feedback

New Scientist•Oct 2, 2024

'A lot of money is going to get wasted': MIT economist sounds the alarm on AI hype

Business Insider•Oct 2, 2024

Learn more about Metaculus NewsMatch

What will be state-of-the-art accuracy on the Massive Multitask dataset on the following dates?

94.9

When Will an AI be Amongst the Best Forecasters on Metaculus

2027-11

When will AI achieve competency on multi-choice questions across diverse fields of expertise (MMLU)?

2025-06

MetaculusM

🚀Metaculus Open-Source Rewrite is Live!

When will an ML model achieve 90% of human-crowd performance on Autocast?

Make a Prediction

What will be state-of-the-art accuracy on the Massive Multitask dataset on the following dates?

When Will an AI be Amongst the Best Forecasters on Metaculus

When will AI achieve competency on multi-choice questions across diverse fields of expertise (MMLU)?

Comments

What will be state-of-the-art accuracy on the Massive Multitask dataset on the following dates?

When Will an AI be Amongst the Best Forecasters on Metaculus

When will AI achieve competency on multi-choice questions across diverse fields of expertise (MMLU)?

What will be state-of-the-art accuracy on the Massive Multitask dataset on the following dates?

When Will an AI be Amongst the Best Forecasters on Metaculus

When will AI achieve competency on multi-choice questions across diverse fields of expertise (MMLU)?

What will be state-of-the-art accuracy on the Massive Multitask dataset on the following dates?

When Will an AI be Amongst the Best Forecasters on Metaculus

When will AI achieve competency on multi-choice questions across diverse fields of expertise (MMLU)?