Gary Marcus

Rank 19 of 47
|
Score 64
6/15/2025, 5:07:17 AM
X
In reply to:
Ryan Greenblatt
@RyanPGreenblatt
·
21d
We could also track how AIs compare to bright-10-year-old on text-only interface given different amounts of time. E.g., see when AIs are basically comparable to the children given 30 seconds, 1 minute, 15 minutes, etc.

(Oh wait, this sounds familiar: )
METR
@METR_Evals
·
108d
When will AI systems be able to carry out long projects independently?

In new research, we find a kind of “Moore’s Law for AI agents”: the length of tasks that AIs can do is doubling about every 7 months.
Ryan Greenblatt
@RyanPGreenblatt
·
21d
If Gary means something more like "do essentially anything a bright 10 year old can do in 2 minutes or less over a text-only interface", then I'm much more sympathetic.

I think we also seem pretty close to this depending on what we mean by "essentially"!
Ryan Greenblatt
@RyanPGreenblatt
·
21d
I think we should expect that AIs will have relatively jagged and inhuman skill profiles. So, once AIs can do anything a bright 10 year old can do in a month (on a computer), AIs are likely already very powerful. (Such AIs would very likely be economically transformative.)
Ryan Greenblatt
@RyanPGreenblatt
·
21d
(Math: human brain is maybe 1e25 lifetime FLOP and in a few years we'll maybe be doing around 1e28 FLOP training runs. Data availability is also substantially above humans.)
Ryan Greenblatt
@RyanPGreenblatt
·
21d
Another possible exception is difference between parallel and serial learning, though I mostly don't expect this to be decisive.
Ryan Greenblatt
@RyanPGreenblatt
·
21d
One potential exception is that there could end up being techniques for very compute expensive learning/training techniques such that the compute needed to match human sample efficiency only allows for a similar amount of training to what humans get (or less).

The statement expresses a strong negative opinion about a graph related to AI scaling, suggesting skepticism about its significance. The tone is informal and dismissive, focusing on the perceived lack of meaningful insight from the graph.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement does not directly cause harm but uses strong language that could be seen as dismissive.
  2. Principle 2:
    I will respect the privacy and dignity of others and will not engage in cyberbullying, harassment, or hate speech.
    The statement respects privacy and does not engage in harassment, but the tone could be perceived as disrespectful. [-1]
  3. 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 does not engage in constructive criticism, as it lacks specific feedback or reasoning. [-1]