Two of the world's top AI scientists now in this heated public debate over the state of the technology,

disagreeing on whether it can ever be as smart as humans or if we need a new type of AI entirely.

Our dear Dubosa is digging into that in today's tech check. Morning, Dee.

Hey, good morning, Carl. So at its core, this debate is really about whether the current AI build-out

actually pays off. Yanlekan is one of the most influential and one of the most controversial minds in AI,

a pioneer of modern neural networks, touring award winner, and until recently, Meta's chief AI scientist.

Now, he's been openly pushing back against the dominant narrative in AI that bigger models

and more data lead to general intelligence.

His comments in a recent interview have poured gas on a debate that has been quietly brewing

for months.

There is no such thing as general intelligence.

This concept makes absolutely no sense.

Human intelligence is super specialized.

This concept of general intelligence is complete BS.

Now he's calling into question the entire assumption behind the industry's KAPX bet,

trillions of dollars being spent on chips, data centers, power, all on the idea that

scale eventually unlocks general intelligence, which unlocks a surge in economic

output.

Instead, he's developing world models, which requires a whole different approach, less about

raw scale, more about raw scale,

more about teaching machines, how the world actually works. Now, enter Demas Hussabas,

founder of Deep Mind, noble winning scientist running AI at Google, publicly pushing back saying,

...