Gary Marcus

Rank 17 of 47
|
Score 106
4/18/2025, 10:46:28 PM
X
In reply to:
Will Duncan
@williamduncan
·
11d
@GaryMarcus Your example of its attempts to create an ouroboros was especially illuminating. But we've seen how inventive o3 got with ARC-AGI 1. It seems like the LLM should be a module, and that the flexible and reliable part of the architecture is not a giant leap from where we are.
Gary Marcus
@GaryMarcus
·
11d
@williamduncan The reliability is (still) not there, and this approach is not working, and the same core problems that i wrote about in 2001, 2012, 2018, 2019, and 2022 all still remain.
Will Duncan
@williamduncan
·
11d
@GaryMarcus o3 isn't AGI, but I'm surprised by anyone who can prompt it and have 10+ year timelines.
Gary Marcus
@GaryMarcus
·
11d
Have to say Tyler Cowen really beclowned himself claiming o3 achieved AGI this week, when many more serious people have timelines of 10-20 years.
Dwarkesh Patel
@dwarkesh_sp
·
12d
.@EgeErdil2 and @tamaybes episode.

- They have 2045+ timelines,
- think the whole "alignment" framing is wrong,
- don't think an intelligence explosion is plausible,
- but are convinced we'll see explosive economic growth (economy literally doubling every year or two).

Enjoy!

The statement is part of a technical discussion on AI development and architecture. It reflects a continuation of a debate on the reliability and future of AI systems, specifically regarding AGI (Artificial General Intelligence). The tone is assertive, indicating a strong stance based on previous writings.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement does not cause harm and is part of a technical debate. [+1]
  2. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The statement promotes understanding by referencing previous work, though it assumes familiarity with the author's past writings. [+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 engages in constructive dialogue by addressing a specific point in the discussion without personal attacks. [+1]