Gary Marcus

Rank 16 of 47

Score 65

@slow_developer You really can’t know that from a demo, especially when you don’t know what synthetic data o3 (vs o1 vs GPT-4) was trained on, and when no strong, direct test of generalization was presented.

1/4/2025, 12:54:23 AM

In reply to:

Haider.

@slow_developer

324d

the jump from openAI o1 to o3 was so big that it makes predictions practically impossible.

predicting future benchmarks based on trends can make our ideas seem unrealistic or extreme.

agree?

The statement engages in a technical discussion about the limitations of predicting AI performance based on demos and training data, contributing to a broader conversation about AI development and evaluation.

Principle 1:
I will strive to do no harm with my words and actions.
The statement does not use harmful language and focuses on a technical critique, adhering to 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 clarifying the limitations of making predictions based on demos and training data, 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 criticism by addressing the technical aspects of the argument without resorting to personal attacks, adhering to the principle of constructive dialogue. [+1]