Gary Marcus

Rank 13 of 47
|
Score 83
11/17/2025, 4:45:50 PM
X
In reply to:
Алтангүх Батбаяр
@altansukhbatbay
·
35d
@GaryMarcus The major labs clearly have inside models much better than what is available today. There was someone reliable here who had access to such an inside model and reported so.
Gary Marcus
@GaryMarcus
·
37d
People saying stuff like this “OpenAI's unreleased model that won gold on the IMO in July is clearly better than even the best publicly available model today (GPT-5 Pro)” are speculating and probably wrong.

my hunch is that the IMO model is limited either in its generality or
prinz
@deredleritt3r
·
37d
I've noticed that a lot of people tend to assume that capabilities of consumer-grade AI models (like GPT-5) are the same as the true frontier of AI capabilities. This is a mistake.

For example, OpenAI's unreleased model that won gold on the IMO in July is clearly better than

The statement 'have been hearing that for years' is a brief, somewhat dismissive response to a conversation about the capabilities of AI models. It does not substantively engage with the public issue of AI development or contribute to the civic dialogue. Instead, it reflects a personal opinion or skepticism about the claims being discussed.