If this is the way to superintelligence, it remains a bizarre one. “This is back to a million monkeys typing for a million years generating the works of Shakespeare,” Emily Bender told me. But OpenAI’s technology effectively crunches those years down to seconds. A company blog boasts that an o1 model scored better than most humans on a recent coding test that allowed participants to submit 50 possible solutions to each problem—but only when o1 was allowed 10,000 submissions instead. No human could come up with that many possibilities in a reasonable length of time, which is exactly the point. To OpenAI, unlimited time and resources are an advantage that its hardware-grounded models have over biology. Not even two weeks after the launch of the o1 preview, the start-up presented plans to build data centers that would each require the power generated by approximately five large nuclear reactors, enough for almost 3 million homes.
Yeah, it always seems that every time someone questions the wisdom or validity of this analogy seems not to understand it.
It’s either the misunderstanding that the constraints of the hypothetical are finite (a million vs. infinity).
Or the insistence that any sufficiently infinitesimal chance is “practically zero”, when literally any likelihood multiplied by infinity is going to guarantee an occurrence.
You can actually expand the infinite monkey theory to say that an infinite number of monkeys using typewriters for an infinite amount of time would write every single book ever written in any language the keyboard is capable of typing, as well as every possible book that could ever even theoretically exist, an infinite number of times, and still be correct.
Any infinite set of random (or even semi-random) characters will contain every possible set of characters that could ever exist, of any length. The works of Shakespeare are also encoded into Pi, we just haven’t calculated enough digits to discover one yet (and very likely never will).