Gary Marcus

Rank 13 of 47
|
Score 97
Gary Marcus
@GaryMarcus
·
67d
“a good set of challenges” for AGI from 2022
• none have been solved
• community accepted them as legit at the time
• goalposts haven’t moved

translation and lessons:
1. people are making shit up when they say I am moving my goalposts; they haven’t changed.
2. those with
7/20/2025, 6:52:31 PM
X
In reply to:
Liron Shapira
@liron
·
67d
@GaryMarcus @lethal_ai @DoomDebates Fwiw I didn’t manage to nail Gary on any major failed prediction when we had our debate :)
Gary Marcus
@GaryMarcus
·
67d
@lethal_ai @liron @DoomDebates in despise this almost enough to block you. here is why - people keep making this accusation and never point to any concrete evidence when i ask. never.
lethalintelligence.ai
@lethal_ai
·
67d
My first thoughts on the “Gary Marcus” test:
(btw @GaryMarcus is great, especially enjoyed the @liron @DoomDebates episode, but I guess I’m just more of a techno-optimist at heart)
Gary Marcus
@GaryMarcus
·
67d
The “Marcus Test” for AGI (I didn’t name it that, but it means “no AGI til Marcus says it’s AGI”) is a pretty good test for AGI.* But don’t bullshit yourself about what it is.

• It’s not about achieving a single result—I have used the word impressive (or similar) to describe

The conversation involves a debate about artificial general intelligence (AGI) and the validity of predictions and standards set by a public figure. It includes elements of disagreement and critique, with references to past statements and challenges. The tone is somewhat contentious, with accusations of moving goalposts and a focus on evidence and accountability.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement and replies engage in a debate about AGI predictions, which can contribute to public understanding, but the tone is somewhat accusatory, potentially causing harm. [-1]
  2. Principle 2:
    I will respect the privacy and dignity of others and will not engage in cyberbullying, harassment, or hate speech.
    There is a risk of disrespecting privacy and dignity, as the conversation involves direct accusations and challenges without clear evidence. [-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 dialogue includes personal attacks and accusations, which detract from constructive criticism and respectful engagement. [-2]