Update 2026-03-06 (PST) (AI summary of creator comment): The creator will resolve this market two weeks after the game's release. Resolves no if no credible evidence of a success case is provided. Key rules for what counts:
A human operator may assist with input, but no coaching during the run
A human operator may update context between failed runs to calibrate the system
The creator plans to personally attempt a few runs with the latest foundation models shortly after release
1,000People are also trading
@qhyzn are you really willing to do that?
A good amount of what this market is measuring is people's willingness to throw AI at solving new recreational games
@2b3o4o I'll resolve this market in two weeks if there is no credible evidence provided of a success case. ( Note: I'll likely attempt a few runs w/ the latest foundation models in a few days if it's not terribly inconvenient to do so ) . For it to count - it's ok for a human to assist with input, however there should be no coaching provided during the run. It is permissable for a human operator to update context after a failed run to calibrate the system on.
Apologies about my earlier comment triggering the summary bot to falsely state that the market will resolve a few days from the release.
@rayman2000 Yes. There are some that aren't bad, but they relied heavily on brute search/simulation strategies, which are largely infeasible a few days after release.
Every LLM approach I've seen so far were terrible