Jonathan Berk

speaker
579 appearances 6 recordings 1 series first heard Jun 2026 last heard 13 Aug

Jonathan Berk’s voice in public audio — every appearance, attributed to the second.

Trend

recordings per month · last 12 months
3 · Jul OctJan 26AprJulnow

Recordings per month over the last 12 months — 6 in all, peaking in Jul 2026 with 3.

Appearances

newest first · ▶ plays the moment
Because then we could ask the question: how similar is AI to the human brain?
And more importantly, if in fact AI isn't so
similar to the brain, how close are we to replacing the human brain?
Which I think is one of the big first order questions in AI.
Yeah, so, you know, a few years ago I read what I consider to be one of the best books on the human brain, written by Jeff Hawkins called A Thousand Brains.
That book starts out with an amazing simple observation, which is that the brain can't be a CPU simply because neurons do not fire often enough to compute.
And so starting with observation, Jeff Hawkins develops this theory of the brain, and it's pretty complex, but at its root, the theory is the brain is a predictive device.
It learns from its environment and based on that predicts what's going to happen in the world and only goes into CPU mode when that prediction doesn't actually occur.
Okay, well, the group.
The great breakthrough I would say for AI was that instead of trying to solve problems, CPU mode, all AI does is predict the next word.
And this similarity led to this incredible breakthrough.
You know, the Turing test, which for those of you who don't know it is, Turing said the definition of an intelligent machine is when a human being interacts with a machine and doesn't know it's the machine.
And that for a very long time was a gold standard in artificial intelligence.
Today it's a joke.
Everybody accepts that you can interact with AI and it looks like a human being.
And so that breakthrough came merely because we stopped trying to solve the problem and all we tried to do was predict the next word.
question I guess is if AI at a very high level is a model of the brain, does this mean that ultimately AI will become the human brain or maybe better than the human brain?
Yeah, I you know, it depends how much important you think architecture is, how much important do you think the approach is.
You know, I don't necessarily share your view on that.
I think the advances in AI that have come just from this predicted model have been so enormous that we obviously must be onto something.
Showing 461–480 of 579 · page 24 of 29 ← Previous Next →