Railroads, electrification, fibre optic networks all lost money for early investors but the infrastructure became the basis for future economic and technological developments.
Asking questions that are absurd is like interviewing developers and asking absurd questions on the grounds that it tests creative and critical thinking.
Remember these Microsoft interview questions designed to identify the best developers?
"If you could eliminate one U.S. state, which one would it be?"
"How would you move Mount Fuji?"
Absurd interview questions have an air of legitimacy due to the quasi sophisticated justifications put forward for why they are good tests.
Absurd interview questions are not good tests of people or LLMs.
> Asking questions that are absurd is like interviewing developers and asking absurd questions on the grounds that it tests creative and critical thinking.
Yes; what's wrong with that?
Do you suppose that it doesn't test those qualities?
I, personally, think it does not. For the absurd task which has no testable solution, there is no grounded basis to judge creativity or critical thinking behind an answer. The only things that remain are "vibes" or "hilarity". Well, if these are what you want to test for, I do not disagree. But it isn't something I'd be interested in.
As I understand it; the point is to ask for an SVG which would demonstrate a conceptual understanding of what is being asked for and that is an important test IMO.
What sufficiently hard, but useful, problem would you ask the model for?
It was never intended as a meaningful benchmark. The surprising thing was that for the first ~12 months performance on the stupid pelican benchmark did seem to correspond to the performance of the models on other tasks.
That pattern no longer holds - Fable 5 and GPT-5.6 have both been out-pelicaned by lesser models now.
reply