The whole resources section is full of zingers, such as this section where they suggest that AI will surpass a bottleneck just by thinking harder
Often, you can just think more and think better, and obviate the need for a test, because you realize that previous observations contain the answer already. Compare the ability of modern AIs to learn how to pilot robots, sometimes using pure simulation.
Sometimes you can think harder until you find a similarly reliable but faster test.
A sufficiently smart reasoner, moreover, also has the option of just figuring out how reality is without needing so many dang experiments. Sometimes the data you already have is enough, if you’re smart enough to interpret it.
They then cite Einstein coming up with the theory of general relativity without mentioning it took him ten years to do so. Meaning even an AI that thinks faster than a human would still take time to think harder