Will LLMs be worse than human level at forecasting when they are superhuman at most things? | Manifold

Will LLMs be worse than human level at forecasting when they are superhuman at most things?

13

1kṀ471

2030

41%

chance

1H

6H

1D

1W

1M

ALL

Resolves to my personal opinion, unless @SemioticRivalry disagrees.

Feel free to ask for clarification about what is meant in the comment but I'll likely keep it vague to resolve to the spirit of the question if precise definitions might take away from that

Get

1,000

to start trading!

People are also trading

Will LLMs mostly overcome the Reversal Curse by the end of 2025?

-4% 1d68% chance

Will LLMs be better than typical white-collar workers on all computer tasks before 2026?

Will LLMs be the best reasoning models on these dates?

Will there be an LLM which scores above what a human can do in 2 hours on METR's eval suite before 2026?

Will an LLM improve its own ability along some important metric well beyond the best trained LLMs before 2026?

At the beginning of 2028, will LLMs still make egregious common-sensical errors?

Will an LLM do a task that the user hadn't requested in a notable way before 2026?

Will LLMs' loss function achieve the level of entropy of human text by the end of 2030?

Will the most interesting AI in 2027 be a LLM?

Will a LLM trained with FP4 have frontier-level performance before 2028?

Related questions

Will LLMs mostly overcome the Reversal Curse by the end of 2025?

Will LLMs be better than typical white-collar workers on all computer tasks before 2026?

Will LLMs be the best reasoning models on these dates?

Will there be an LLM which scores above what a human can do in 2 hours on METR's eval suite before 2026?

Will an LLM improve its own ability along some important metric well beyond the best trained LLMs before 2026?

At the beginning of 2028, will LLMs still make egregious common-sensical errors?

Will an LLM do a task that the user hadn't requested in a notable way before 2026?

Will LLMs' loss function achieve the level of entropy of human text by the end of 2030?

Will the most interesting AI in 2027 be a LLM?

Will a LLM trained with FP4 have frontier-level performance before 2028?

© Manifold Markets, Inc.•Terms•Privacy