It's easy to envision one or two purely-AI-written books, given good marketing, making it onto the NYT fiction bestseller list in the capacity as novelties. Therefore, if we want to gauge proliferation of well-written AI-written fiction capable of appealing to audiences not primarily seeking novelty, we want to check for a larger count. I have, for this market's purposes, arbitrarily chosen a target of 5. So. Do we expect at least five purely-AI-written books to make it onto the NYT fiction bestseller list before the end of 2030?

(This could, of course, be confounded by e.g. the NYT establishing a policy of not putting AI-written novels onto the list even if they're selling well. If there exist better benchmarks for the same phenomenon, I encourage others to make similar markets resolving against those better benchmarks and link them in the comments here.)

'purely-AI-written' here means books whose text is generated and, if applicable, edited by AI with little to no human involvement, with—let's say, as a theoretical concrete benchmark, although it'd be hard to check most of the time in practice—less than 1% of the final word-count consisting of text either written or edited-into-its-current-form by a human.

I don't actively monitor the NYT bestseller list or its contents, and so I will rely on market-participants to bring news of any relevant novels should they be released. I will, after market closing time, do a bit of research myself in case any relevant books showed up on the list without comment, prior to resolving one way or the other.