Gary Marcus

Rank 21 of 47
|
Score 55
4/20/2025, 9:03:38 PM
X
In reply to:
Nathan Labenz
@labenz
·
71d
@GaryMarcus I doubt we will agree on the scoring tbh

I would give you 4-5/7

- reasoning, multi modality, voice mode, and long context do constitute a massive advance imo - certainly my utility is way up!

- leaders have major moats! for starters, in compute infra, but lots more. That’s
Gary Marcus
@GaryMarcus
·
71d
@labenz Did you beat these?
Gary Marcus
@GaryMarcus
·
477d
Prediction: By end of 2024 we will see
• 7-10 GPT-4 level models
• No massive advance (no GPT-5, or disappointing GPT-5)
• Price wars
• Very little moat for anyone
• No robust solution to hallucinations
• Modest lasting corporate adoption
• Modest profits, split 7-10 ways
Nathan Labenz
@labenz
·
71d
@GaryMarcus While I wouldn’t want to trade prediction records from the last 24 months, I do recognize that you earn some Bayes points here

Notably o3 represents a pretty dramatic trend reversal on hallucinations - I wonder if that signals the start of a new trend or will prove a blip
Gary Marcus
@GaryMarcus
·
71d
@labenz Who could have predicted this, @labenz?
Nathan Labenz
@labenz
·
74d
Making up lovely AirBnB host details really limits o3's utility as a travel agent

At least it came clean when questioned I guess? 🤷‍♂️🙃

The statement '@labenz lol' is a lighthearted, informal response that does not engage substantively with public issues or debates. It appears to be a casual, humorous reply within a conversation about AI predictions and advancements.