INSIGHTS

IDEAS, METHODS, AND TECHNOLOGY SHAPING BETTER FORECASTING AND STRATEGIC DECISIONS.

2024-09-18The Forecasting Machine

AI versus the crowd: Hypermind’s hybrid forecast map of the 2024 US presidential election

In September 2024 Hypermind set its new retrieval-augmented AI Forecasting Machine against its human prediction market on a state-by-state election map. How the experiment worked, what it forecast, and what the result can and cannot tell us.

2023-09-20The Forecasting Machine

“Governing is predicting”: crowd forecasting at the Brussels Democracy Lab

At a Democracy Lab on smarter crowdsourcing in Brussels on 19 September 2023, Émile Servan-Schreiber gave EU policymakers five reasons to use crowd forecasting, starting with the idea that governing is predicting.

2022-10-21The Forecasting Machine

Do prediction markets reward the lone right answer? Servan-Schreiber’s reply to a PNAS model of collective intelligence

A 2017 PNAS model found that “market rewards” make crowds herd and proposed paying accurate minorities instead. Émile Servan-Schreiber argues that real prediction markets already pay that way.

2022-07-11The Forecasting Machine

“Arising Intelligence”: what Hypermind’s crowd got right and wrong about AI progress

From 2021, Hypermind ran contests asking forecasters to predict AI benchmarks with Jacob Steinhardt of UC Berkeley. Progress on maths and language beat the crowd’s expectations; robustness lagged them.

2022-05-02The Forecasting Machine

Bettors versus pollsters: the Hypermind–Le Point panel and the 2022 Macron–Le Pen run-off

Across 143 daily rolling polls of the 2022 French presidential run-off, the Hypermind–Le Point panel’s forecast was closer to the final result in 128 cases. What was compared, how the panel worked, and where the comparison has limits.

2022-01-11The Forecasting Machine

Ask what your circle thinks: an implicit super-sample for surveys and forecasts

A 2022 Electoral Studies paper examined why asking about other people's preferences can outperform asking only about one's own—and how the same mechanism may help prediction markets.

2021-11-21The Forecasting Machine

What 15 months of crowd forecasting taught Johns Hopkins about disease outbreaks

A study by the Johns Hopkins Center for Health Security and Hypermind, published in November 2021, followed 562 forecasters through 61 questions on 19 diseases. The combined crowd forecast was well calibrated, beat chance by a wide margin and outscored every individual, including the public-health experts.