There was this article I read about crowd decision making, but I cannot for the life of me recall where this tidbit was in, but I found the conclusion interesti

There was this article I read about crowd decision making, but I cannot for the life of me recall where this tidbit was in, but I found the conclusion interesting. Essentially wisdom of the crowds works when the members have at least some knowledge of the underlying thing. The example I heard first was guessing the weight of a bull at an auction, where the averaged crowd estimate was incredibly close to the actual weight, more so than any of the local expert estimates. The Oscars cited in this article would be another example of something where members of the crowd have some knowledge of the event, and the aggregate knowledge may surpass expert knowledge. But then there's the same kind of crowd predictions in things crowd members have little to no knowledge of, where things frequently deviate from actual results. Crowd expert systems will fail at things like medical diagnosis or aircraft maintenance which require high degrees of specialized knowledge the average person possessed little of. Applying them to systems like these results in horrible mistakes happening. If anyone does recall the book or article this was in, that'd be neat, even though I think the point stands to a degree. Crowd intelligence isn't necessarily the panacea it is made out to be.