Recursive forecasting: Eliciting long-term forecasts from myopic fitness-seekers — AI Alignment Forum

We’d like to use powerful AIs to answer questions that may take a long time to resolve. But if a model only cares about performing well in ways that…

x Recursive forecasting: Eliciting long-term forecasts from myopic fitness-seekers — AI Alignment Forum Outer Alignment Redwood Research AI Frontpage 29 Recursive forecasting: Eliciting long-term forecasts from myopic fitness-seekers by Jozdien , Alex Mallen 28th Apr 2026 9 min read 2 29 We’d like to use powerful AIs to answer questions that may take a long time to resolve. But if a model only cares about performing well in ways that are verifiable shortly after answering (e.g., a myopic fitness seeker ), it may be difficult to get useful work from it on questions that resolve much later. In t

saved by

• Emanuel Ruzak

related reading

• The behavioral selection model for predicting AI motivations — LessWronglesswrong.com
• The AI Superforecasters Are Here - by Scott Alexanderastralcodexten.com
• Self-Fulfilling Misalignment Data Might Be Poisoning Our AI Modelsturntrout.com
• The Unreasonable Difficulty of Time Series Forecastingsuzyahyah.github.io
• Fitness-Seekers: Generalizing the Reward-Seeking Threat Model — LessWronglesswrong.com
• Without specific countermeasures, the easiest path to transformative AI likely leads to AI takeover — LessWronglesswrong.com
• What 2026 looks like — AI Alignment Forumalignmentforum.org
• Will reward-seekers respond to distant incentives? — LessWronglesswrong.com
• Training LLMs to Predict World Events (Guest Post with Mantic) - Thinking Machines Labthinkingmachines.ai
• Import AI 455: AI systems are about to start building themselves.importai.substack.com
• Pitfalls in Evaluating Language Model Forecastersarxiv.org
• Risk from fitness-seeking AIs: mechanisms and mitigationssubstack.com