Similar with shorter timeframe:

The phrase "reputable journal" is doing a lot of heavy lifting here. What counts as "reputable"?

@apetresc market creator has a bet on no, so I’m guessing it will be an arbitrarily high bar.

@CoryS I bet NO because I predict it will happen in 2027 or later. I will write something to clarify the "reputable journal" phrase. (One would want the journal to be Q1-Q2, indexed by Scopus or similar, etc.)

@apetresc I'm far more scared by 'co-author' ... plenty of current papers have had AI's being used in the research

@AdamTreat I meant 'co-author' in the exact sense that would make people scared, lol.

@ken Can you please add some criteria on what counts and what doesn't? For example:

If a human brute forced a thousand cases in some problem, than he would be a co-author (I think that would also happen if he were guy who coded the brute force). If AI bruteforcing also counts, than it already happenad decades ago.

Some journals forbid mentioning AI as coauthors. So how would we know it even happened?

This is absolutely going to happen. It's not AI, but some eccentric people have been listing computers as coauthors for a long time, see for example https://sites.math.rutgers.edu/~zeilberg/ekhad.html