Alex Wiltschko

Like you're out of data.

You're out of the Internet.

It's gone.

We've digested it all.

And I don't think that we've seen the amount of crazy progress in the past year that we saw, say, you know, year five ago to year four ago.

Like that was wild progress.

Right.

And now things are getting better and we're adding new features, but we're adding features to products.

We're not growing some capability nearly as fast as we were before.

This is my hot take, which is language models are great and all, but when we talk about progress in AI in general, we're actually talking about dozens of different technologies.

And so what I'm seeing is a bit of a smoke and mirrors or masking game, which is like, oh, the progress is the same rate.

No, no, no, no.

Progress on video generation is a different S-curve.

Progress on image generation is a different S-curve.

Now, you can build products to integrate all this stuff.

Oh, my gosh, can you create so much value in society by combining these things to products?

It's not the same as technological progress, right?

So if what you want is to figure out when your AI god will wake up, I don't agree with that.

I don't think that is a thing.

I think what we have is a new form of steam turbine engine, but for mental work.