Gavin Baker

speaker

289 appearances 1 recordings 1 series first heard Aug 2025 last heard Aug 2025

Gavin Baker’s voice in public audio — every appearance, attributed to the second.

Trend

recordings per month · last 12 months

No recordings in the last 12 months.Older appearances are listed below; set an alert to hear about the next one.

newest first · ▶ plays the moment

problem in AI is what's called benchmark saturation, which is just these models are getting so smart.

That benchmarks that were optimized even for, you know, human graduate students are not sufficient to measure their intelligence.

So we've developed these new benchmarks.

There's humanities last exam on the left.

And you can s can you

It is a list.

I everybody you could just Google it, but I mean the questions are ridiculously hard.

They're extremely hard questions from a wide range of disciplines.

And you can see Groc 4 when it came out and it has improved.

Since since its release roughly a month ago, was at 44.4%.

And GPT-5 was at 42.

And similarly, RKGI2, which is another benchmark.

Or something called the jagged frontier in AI, where things that are very easy for AIs, like maybe math.

Are really hard for humans and vice versa.

And RKGI measures tasks that are relatively easy for humans.

Maybe you can almost think of, you know, there's a lot of like geometric puzzle solving, but very hard for AIs.

And here again, GROK 4 is, you know, significantly better than GPT-5.

Now, in fairness, and you know, to be balanced and accurate, it is narrowly, very narrowly ahead in artificial analysis on the right.

But this is this is the first time they have not released a model that wasn't decisively better.

And the progress for Groc Four was exceptional.

Showing 21–40 of 289 · page 2 of 15