Sebastian Raschka

speaker

1,024 appearances 1 recordings 1 series first heard Feb 2026 last heard 1 Feb

Sebastian Raschka’s voice in public audio — every appearance, attributed to the second.

Trend

recordings per month · last 12 months

Recordings per month over the last 12 months — 1 in all, peaking in Feb 2026 with 1.

Appearances

newest first · ▶ plays the moment

Lex Fridman Podcast · #490 – State of AI in 2026: LLMs, Coding, Scaling Laws, China, Agents, GPUs, AGI · 1 Feb 2026 podcast

And then after the fact, someone revealed that.

But AlphaFold, when it came out, it crushed this benchmark.

There are also multiple iterations.

But I remember the first one, I'm not an expert in that subfield, but the first one explicitly modeled the physical interactions of the physics of the molecule, also like the angles, impossible angles.

And then in the next version, I think they got rid of this.

And just with brute force scaling it up.

And I think with LLMs, we are currently in this brute force scaling because it just happens to work.

But I do think also at some point it might make sense to bring back this...

And I think with world models, I think that might be actually quite cool.

And of course, also for robotics.

That is completely unrelated from LLMs.

I think it's also related to the continual learning issue, essentially, where the real world is so complex where with LLMs, you don't need to really have something learned for the user because there are a lot of things everyone has to do.

Everyone maybe wants to fix their grammar in their email or whatever.

code or something like that it's more constrained so you can kind of prepare the model for that but preparing the robot for the real world that's harder i mean you have the foundation models the robotic foundation models but you can learn certain things like grasping things but then again i think everyone's house is different you know like it's so different and that is i think where the robot would have to learn on the job essentially and i think that i guess is the the bottleneck right now like how to you know customizing it on the fly essentially

I think what's interesting is to think about whether these systems will be independent, like completely independent in the sense that, well, I have no doubt that LLMs will kind of at some point solve coding in a sense like calculators solve calculating, right?

So at some point humans developed a tool that, you know, you never need a human to calculate that number.

You just type it in and it's an algorithm.

You can do it in a

in that sense and i i think that's the same probably for coding but the question is so i think what will happen is yeah you will just say build that website it will make a really good website and then you maybe refine it but will it do things independently where so will you be still having humans asking the ai to do something like will there be a person say build that website or will there be ai that just builds websites or something or whatever i think

And I think this gets to the point that you mentioned that some people you talk to are skeptical.

Showing 781–800 of 1,024 · page 40 of 52