Gary Marcus

Rank 14 of 47

Score 83

@slow_developer First one, clearly. if that wasn’t clear before surely it should be obvious after Orion got downgraded to 4.5.

Option two has been amply tested and should be rejected; option 3 has no compelling argument.

2/13/2025, 2:53:09 PM

In reply to:

Haider.

@slow_developer

340d

there are 3 camps divided over AGI:

- it will not emerge from LLM

- its properties will emerge once LLMs reach a critical mass (e.g., 1000 trillion parameters)

- it will never emerge since consciousness is non-algorithmic and not purely physical

which one is more appropriate?

The statement engages in a discussion about the emergence of Artificial General Intelligence (AGI) from large language models (LLMs), which is a substantive public issue in the field of artificial intelligence and technology. The tone is analytical and evaluative, as it critiques different perspectives on AGI.

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 evaluating the options presented in the original conversation. It does not engage in personal attacks or ad hominem arguments, focusing instead on the merits of the arguments themselves. [+1]