Gary Marcus

Rank 22 of 47
|
Score 36
Gary Marcus
@GaryMarcus
·
455d
@AISafetyMemes Yes there are more tools; integration with the tools has been problematic though. I certainly think that integrating w symbolic tools would be good if it can be done. But so far that has been shaky.

You can’t just say you personally don’t care about hallucinations; that is a…
6/18/2024, 2:15:51 PM
X
In reply to:
poor black box
@PoorBlackBox
·
455d
@AISafetyMemes @GaryMarcus Voicing agreement to this, and that a response to this would be useful @GaryMarcus -- it captures my disagreements with your perspective and I predict many others too
AI Notkilleveryoneism Memes ⏸️
@AISafetyMemes
·
455d
@GaryMarcus 1: Yes, but the trend is clear that more compute = smarter models. GPT4 is far more capable than a high school student in the ways that I care about.

I don't care that you can trick it with logic puzzles more often (which I'm not even entirely sure is true) or that it…
Gary Marcus
@GaryMarcus
·
455d
Soooo very wrong, in so many ways
1. That chart is make-believe, with wacky y-axes and totally hypothetical data. And no GPT-4 is not really a substitute for a smart high school student, c’mon.
2. Not much progress on raw capabilities since GPT-4 (trained Aug 2022, almost two…
AI Notkilleveryoneism Memes ⏸️
@AISafetyMemes
·
455d
The same pundits have been saying "deep learning is hitting a wall" for a DECADE.

Why do they have ANY credibility left?

Wrong, wrong, wrong. Year after year after year.

Like all professional pundits, they pound their fist on the table and confidently declare AGI IS DEFINITELY…

The statement and the preceding conversation constitute public discourse as they engage in a substantive debate about the future of artificial intelligence (AI), specifically the potential for achieving artificial general intelligence (AGI) and the implications of scaling laws in AI development. The conversation involves multiple participants discussing the validity of predictions and the interpretation of data related to AI progress.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The conversation generally strives to do no harm, focusing on technical and theoretical disagreements rather than personal attacks. However, some statements, such as 'Soooo very wrong, in so many ways,' could be perceived as dismissive and may not fully uphold this principle. [-1]
  2. Principle 2:
    I will respect the privacy and dignity of others and will not engage in cyberbullying, harassment, or hate speech.
    The conversation respects the privacy and dignity of others, as it does not engage in cyberbullying, harassment, or hate speech. The focus remains on the topic of AI development. [+1]
  3. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The conversation promotes understanding and empathy to some extent by engaging in a dialogue about differing perspectives on AI progress. However, the tone of some responses could be more constructive to foster better understanding.
  4. 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 conversation involves constructive criticism and dialogue, but there are instances where the tone could be improved to avoid personal attacks or ad hominem arguments. For example, 'Soooo very wrong, in so many ways' could be rephrased to focus on the specific points of disagreement rather than dismissing the other person's perspective outright. [-1]