Infinite Monkeys

I watched this video today and it got me thinking about the infinite Monkey theorem.

The theorem

The infinite monkey theorem states that a monkey hitting keys independently and at random on a typewriter keyboard for an infinite amount of time will almost surely type any given text, including the complete works of William Shakespeare.[a] More precisely, under the assumption of independence and randomness of each keystroke, the monkey would almost surely type every possible finite text an infinite number of times. The theorem can be generalized to state that any infinite sequence of independent events whose probabilities are uniformly bounded below by a positive number will almost surely have infinitely many occurrences.

back to the talk

The video, essentially states, that rather than thinking about integrating AI functionality into our business applications we should think about integrating our business logic into the AI capabilities.

In the diagram, it shows transitioning apps from hand crafted to code + verifier.

The demo

It goes on to demonstrate a completely LLM generated UI for email navigation.

It was bug ridden, it was slow, it was kludgy.

But it gets at the future, ceding complete control to this 'intelligence'.

This brings me back to the infinite monkey.

We're at the point where just throwing compute, more compute, and more compute at the problem will solve problems.

Back to the monkies

We've trained monkeys. They're well-read monkeys and they're reading more.