If we conduct evolution on AI forecasters in a completely AI-based prediction ones—deleting ones that lose all money, and allowing ones that succeed to clone and mutate—will we get significantly better forecasters? Is this faster than doing this on a real prediction market because of how reactive we can make LLMs to the market, and because we can make way more markets?
If you are inspired by this idea, you can reach out to the authors for collaboration or cite it:
@misc{holtzman-evolving-forecasters-2026,
author = {Holtzman, Ari},
title = {Evolving Forecasters},
year = {2026},
url = {https://hypogenic.ai/ideahub/idea/NDwuScGxjj0DdYP3gRTG}
}Please sign in to comment on this idea.
No comments yet. Be the first to share your thoughts!