Remember all that fantasy-talk about 10x-ed productivity and billion dollar unicorns with a single employee? A dose of reality: “AI has not yet obviously translated into faster growth for YC startups
@garymarcus
-
AI’s Growing Negative Perception Despite Positive Potential
By
–
What I hate, truly hate, is that AI (which I love, in principle) has started to feel the same way. Dirty. https://
x.com/JoeAIfollower/
/JoeAIfollower/status/1964738498998734949
… -
Big Tech Questions GenAI ROI and Implementation Costs
By
–
The big companies finally figured it out. GenAI sounds cool but doesn’t work well enough to justify the costs. Chart from @ApolloAcademy
-
Do GenAI Benefits Outweigh the Costs?
By
–
on balance, have the benefits of genAI outweighs the costs? feel free to explain your answer in the comments
-
OpenAI Papers: More Marketing Than Actual Research?
By
–
Has there been an OpenAI paper this decade that was NOT more PR than research?
-
Will AI Make Revolutionary Discoveries by 2030?
By
–
will still be making discoveries like this in 2030? https://t.co/BZBXlxLAff
— Gary Marcus (@GaryMarcus) 6 septembre 2025will still be making discoveries like this in 2030?
-
AGI Timeline Challenges: Core Cognitive Problems Remain
By
–
8-15 and we might reach an agreement. 4 still seems awfully unlikely to me. to many core cognitive problems aren’t really being addressed, and solutions may take a while to roll once we find the basic insights we are lacking. but it’s a fair question.
-
Deepfake video of tech figure spreads without verification
By
–
There is no evidence that this poster below realized that the video she was sharing was fake.
— Gary Marcus (@GaryMarcus) 5 septembre 2025
It’s not even a particularly good fake (soundtrack is pretty far off from the lip movements, and it’s wrong on a bunch of facts, and certainly not what Gates actually would have said,… https://t.co/gfiPokux6NThere is no evidence that this poster below realized that the video she was sharing was fake. It’s not even a particularly good fake (soundtrack is pretty far off from the lip movements, and it’s wrong on a bunch of facts, and certainly not what Gates actually would have said,
-
ChatGPT loves me but it is very much unrequited
By
–
ChatGPT loves me. but it is very much
unrequited -
LLMs accountability reliability hype limitations challenges
By
–
don’t insult my new friend @Lauren_79
;
it is not a skill issue. it is a hype and reliability issue, exacerbated by the fact that LLMs don’t know what they don’t. better prompts might sometimes yield better results. sometimes. but they should be held responsible for the output