Collective Intelligence: Group Decision-Making Benefits

Who created and checked this page

Correction history

Every accepted correction to this page is recorded with the exact change, so readers can see how the page improved over time.

1. 28 July 2026 · corrected by Emir Baycan

   Qualified three overstated claims: softened ant-colony 'globally optimal foraging paths' to 'highly efficient, near-optimal' (not a provably global optimum); qualified 'prediction markets consistently outforecast individual experts' to 'frequently,' noting track record varies by market; and reframed Tetlock's forecasting-tournament findings as showing structured crowds can match or outperform many experts in specific tournaments rather than a universal law that diverse crowds always beat expert consensus. The Lorenz 2011 conformity study, the Physarum/Tokyo-rail-network experiment, and the diverse-teams/ensemble-methods claims were accurate and left in place.

   Before

   The article said ant colonies produce globally optimal paths, prediction markets consistently outperform experts, and Tetlock established a universal finding that crowds outperform expert consensus.

   After

   Qualified each claim to what the underlying research supports (near-optimal rather than global optimum; frequent rather than consistent outperformance; tournament-specific rather than universal).

   Why: These were reasonable summaries stated more absolutely than the research supports. Corrections add the appropriate qualifications rather than adding new sources.

   View the full record →

Full contribution timeline

1. Correction 28 July 2026

   Emir Baycan: Qualified three overstated claims: softened ant-colony 'globally optimal foraging paths' to 'highly efficient, near-optimal' (not a provably global optimum); qualified 'prediction markets consistently outforecast individual experts' to 'frequently,' noting track record varies by market; and reframed Tetlock's forecasting-tournament findings as showing structured crowds can match or outperform many experts in specific tournaments rather than a universal law that diverse crowds always beat expert consensus. The Lorenz 2011 conformity study, the Physarum/Tokyo-rail-network experiment, and the diverse-teams/ensemble-methods claims were accurate and left in place.

   Proof record →