Gary Marcus

Rank 17 of 47
|
Score 71
6/28/2025, 7:42:09 PM
X
In reply to:
Lawrence Chan
@justanotherlaw
·
55d
@GaryMarcus @ohabryka Top 1%* gah
Lawrence Chan
@justanotherlaw
·
55d
@GaryMarcus @ohabryka Peter Wildeford is one of the top forecasters in the world, in that he consistently scores within the top 0.1% of forecasting tournaments and is ranked 5-7 out of 30k+ on Metaculus:

https://www.metaculus.com/accoun…
Gary Marcus
@GaryMarcus
·
55d
@ohabryka show me the data.
Oliver Habryka
@ohabryka
·
55d
@GaryMarcus A random list of people with better track records:

Tamay Besiroglu, Ege Ergil, Peter Wildeford, Eliezer Yudkowsky, Demis Hassabis, Paul Christiano, Daniel Kokotajlo, Eli Lifland, Thomas Larsen.

Probably many more, but would need to look up names.
Gary Marcus
@GaryMarcus
·
55d
@ohabryka elsewhere you have made an argument (nitpicky in my view) challenging a few of the 7 predictions.

to the extent that you think my arguments are “exceptionally bad”, if you applied the same standards as you applied to me, who would be better? “exceptionally bad” is a very strong
Oliver Habryka
@ohabryka
·
56d
@GaryMarcus I… do not believe you have anything close to a 7/7 prediction track record for AI prediction in 2024. Indeed your predictions have so far struck me as exceptionally bad compared to most people trying to forecast AI.

The statement is part of a broader discussion on the credibility and track records of individuals in AI forecasting, which is a public issue related to the development and understanding of artificial intelligence. The tone is somewhat critical, questioning the consistency of AGI timeline predictions by Eliezer Yudkowsky and Demis Hassabis. The intent seems to be to scrutinize the reliability of these figures' predictions, which is relevant to public discourse on AI forecasting.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement does not directly cause harm but implies criticism of individuals' forecasting consistency, which could be seen as undermining their credibility. [-1]
  2. Principle 4:
    I will engage in constructive criticism and dialogue with those in disagreement and will not engage in personal attacks or ad hominem arguments.
    The statement engages in criticism but does not provide constructive dialogue or evidence to support the claims, and it could be perceived as a personal attack on the individuals' credibility. [-2]