Gary Marcus

Rank 12 of 47
|
Score 83
9/18/2024, 12:53:23 PM
X
In reply to:
Prof Les Carr
@lescarr
·
491d
Ironically, LLMs being demonstrably bad at arithmetic only strengthens the argument that they exhibit human-like intelligence. Turing's very first example of his test was the computer failing a task "Q: Write me a sonnet. A: Count me out on this one. I never could write poetry."
Gary Marcus
@GaryMarcus
·
491d
Is 53% “decent” accuracy? Asking for a calculator friend.

Also: cool new study that (roughly speaking) highlights the difference between interpolation (green) and extrapolation (red). A calculator would be green in the whole grid; both GPT systems eventually fall apart.

The statement by @lescarr critiques the Turing Test as a measure of intelligence, suggesting it is more indicative of human gullibility. This is a substantive engagement with the topic of artificial intelligence and its evaluation, contributing to the public discourse on the effectiveness of the Turing Test.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement does not cause harm and is a critique of a concept rather than an individual, adhering to the principle of doing no harm. [+1]
  2. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The statement promotes understanding by challenging the validity of the Turing Test and encouraging a deeper examination of what constitutes intelligence, aligning with the principle of promoting understanding, empathy, and compassion. [+2]
  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 criticism of the Turing Test without resorting to personal attacks, adhering to the principle of engaging in constructive criticism and dialogue. [+1]