This market will resolve to the first 50% time horizon, as reported by METR, for Grok 4.2 / Grok 4.20. If Grok 4.2 is never released, this market resolves N/A. If there is some coding- or other form of specialized version (e.g. Grok 4.20 Code), it will not count for the purpose if this market. A model expected to be called Grok 4.20 Heavy (using parallel test-time compute) would not count for this market, but Grok 4.20 Thinking, Grok 4.20 High, or Grok 4.20 xhigh, would all count for the purpose of this market.
50% time horizon is a measure of AI autonomy based on the length of tasks that AI can do: roughly, it is the time that humans take to complete tasks that an AI system can successfully do 50% of the time. See METR's "Measuring AI Ability to Complete Long Tasks" for the technical definition. Claude 3.7 Sonnet, released in February 2025, was the leading model with a 50% horizon of 59 minutes.
Left bounds inclusive, right bounds exclusive.
See also:
/Bayesian/gpt-52-pro-metr-time-horizon
/Bayesian/gemini-3s-50-time-horizon-per-metr
/Bayesian/grok-420s-metr-50-time-horizon (this market)
/Bayesian/claude-sonnet-46s-metr-50-time-hori
/Bayesian/grok-5s-50-time-horizon-per-metr
1,000The members of the AI futures project have given an update and they appear to now be relying on the 80% time horizon length graph from METR for their predictions rather than the 50% time horizon length graph. This implies that a 50% time horizon is not enough. While I think markets for 50% time horizons are useful, I now think that more attention needs to be paid to 80% time horizon lengths. I am planning to create markets for 80% time horizons either tonight or some other time this week unless someone beats me to it.