Pete Huang

So what he's saying is 1 million to 100 million times more.

Then he's looking forward and estimating from GPT-4 today onwards, we're gonna get another five to six orders of magnitude.

So that's another 100,000 times to 1 million times improvement.

And parts of that are somewhat believable, right?

Like Microsoft is already saying quite publicly that they helped build an absolutely massive computer to help OpenAI train GPT-5.

And people are already calling for even more computing resources, even larger,

computing clusters to train these models.

Ultimately, Leopold looks at all this and comes to the conclusion saying, look, the train isn't stopping here.

I mean, each of these three buckets still has lots of room to run.

And honestly, I believe that 1 million times more training starting from GPT-4 will turn into a human level AI researcher.

And again, once you get one AI that can be an AI researcher, you have millions, and now you're on a path to super intelligence.

Before we continue, let's just talk very quickly about why AGI and ASI are so important.

Once you have AI that is powerful enough,

You can do so many things that you can't even think about today.

Let's talk about weapons.

Any government that gets superintelligence can make crazy powerful weapons.

They can hack into any system.

I mean, the limits of what we think are possible are defined by human abilities.

And now we have AI that can do more than humans.

I mean, definitely for knowledge workers like me who spend our days just typing on a laptop all day.