Gary Marcus

Rank 19 of 47
|
Score 64
6/29/2025, 4:23:08 PM
X
In reply to:
Vojtech Müller
@cobalamin_12
·
10d
@chatgpt21 @GaryMarcus We simply aren't.
GPT-5 won´t be noticeably better than o3.
Veo4 won't be noticeably better than Veo2.

We are no closer to solving LLM´s fundamental shortcomings than we were three years ago.
Chris
@chatgpt21
·
10d
@GaryMarcus Gary, when I said it’s about the rate of progress, I meant it. Progress isn’t linear but that doesn’t mean it’s not progress. One step back and two forward is still net gain. O3 hallucinates less than 3.5. Zoom out over three years and it’s obvious: we’re flat out accelerating.
Gary Marcus
@GaryMarcus
·
10d
You think “it’s all about the rate of progress”?

Ok, let’s be real. Progress in AI is not uniform.

• Veo3’s graphics are far better than models of a couple years ago. [enormous progress]

• But o3 hallucinates more than some of its predecessors. [negative progress]

• And

The statement is part of a public discourse on the progress and development of AI technologies, specifically comparing the advancements in graphics and language models. It engages in a substantive discussion about the capabilities and limitations of AI models, contributing to the broader debate on technological progress.

  1. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The statement promotes understanding by acknowledging the potential for improvement in graphics and the uncertainty regarding language model advancements. [+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 constructive dialogue by responding to a previous comment with a reasoned perspective, without resorting to personal attacks. [+1]
  3. Principle 7:
    I will uphold the principles of free speech and use my platform responsibly and with integrity.
    The statement upholds free speech by contributing responsibly to the discussion on AI progress. [+1]