Will any agent perform better on Minecraft (or comparable open world game) after being fine-tuned on a manual by 2027?
One-OffAI1moManifold MarketsNo KYCVerified resolution dataWell calibratedCurrent community forecastYes 79.8%
Forecasters
17
Question type
binary
Methodology
Play-money forecasting platform
Source type
Forecast
Market data
Updated 10 hours ago
StaleJul 11, 22, 7:56 PM UTCJan 1, 27, 5:00 AM UTC
Trends
Related Markets
Related questions for context, not a comparison of equivalent contracts.
Will Annie Andrews perform better than Jaime Harrison did in 2020, in the 2026 US Senate election in South Carolina?
Manifold MarketsWill an ai be trained/fine-tuned on Quora by 2035?
Manifold MarketsJumanji: Open World · Rotten Tomatoes score
KalshiConditional on the world being a simulation, what is the probability that you're the only conscious entity in it?
Manifold MarketsWill Nintendo release an open world Legend of Zelda-related game made specifically for VR by the end of 2030?
Manifold MarketsLargest IPO by market cap in 2026?
PolymarketRules
To clarify: the experiment is that there are two copies of an agent that runs on Minecraft (or some other open world game environment).
- The agent has the capacity to be fine-tuned with text.
- One version is passed a manual for the game as text (or text + images, but *not* video), the other runs without any finetuning.
- Will the former perform better than the latter (either better sample efficiency or better final reward)?
- The agent can't have been trained on that env before, but it can be trained on other envs/data beforehand (e.g. it's okay if there's a pretrained LLM in the loop).