Gary Marcus

Rank 14 of 47

Score 83

@XiXiDu Good on for you knowing the word subitizing!

It’s precisely for these sorts of reasons, though, that we invented and mainstreamed symbolic algorithms that we implement digital computers. Nobody runs a database on a person, and nobody should run one on an LLM, either.

Since we

6/7/2025, 1:21:30 PM

In reply to:

Alexander Kruel

@XiXiDu

202d

In a rational world, this important work would be the final nail in the coffin of human reasoning:

1. Human working memory has a finite capacity. When the amount of information required to understand and solve a problem exceeds this capacity, learning and performance collapse.

Gary Marcus

@GaryMarcus

204d

Breaking New Results.

In a rational world, this important work would be the final nail in the coffin of LLMs

and so much nonsense about “reasoning models”

People may choose to ignore it, for a while, but the writing is on the wall.

The statement engages in a discussion about the limitations of human reasoning and the use of symbolic algorithms in digital computers, in response to a conversation about reasoning models and their capabilities. It suggests that databases should not be run on humans or LLMs, implying a critique of current AI models.

Principle 1:
I will strive to do no harm with my words and actions.
The statement does not cause harm and contributes to a technical discussion, aligning with the principle of doing no harm. [+1]
Principle 3:
I will use my words and actions to promote understanding, empathy, and compassion.
The statement promotes understanding by discussing the role of symbolic algorithms in computing, aligning with the principle of promoting understanding. [+1]
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, aligning with the principle of engaging in constructive criticism. [+1]