Gary Marcus

Rank 24 of 47
|
Score 25
9/2/2025, 1:45:07 PM
X
In reply to:
Marco Pfeiffer
@TheTrueNemo
·
14d
@GaryMarcus @peterwildeford They do have a 80% accuracy version: https://metr.org/blog/2025-03-19…

I don't know what to make of this test since the result is heavily dependent on how clear the goal is and what tools are available to the llm.
Gary Marcus
@GaryMarcus
·
15d
@peterwildeford do you really think gpt-5 can do anything a human can in 4 hours with 50% accuracy?

and what would the curve be for 100% accuracy?

and what is the value of 50% accuracy?
Peter Wildeford🇺🇸🚀
@peterwildeford
·
15d
"GPT-5 is on trend! GPT-5 is on trend!! deep learning isn't yet hitting a wall!", i continue to insist as i slowly shrink and transform into a corn cob

The statement questions the timeline and performance improvements of GPT-5, engaging in a technical discussion about AI capabilities. It reflects skepticism about the rapid advancements and accuracy claims of AI models.

  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 inquiry. [+1]
  2. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The statement promotes understanding by questioning and seeking clarification on AI advancements. [+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 questioning the claims made about AI performance. [+1]