#预测市场 The prediction market is currently facing a real challenge: AI-generated fake public opinion. Imagine the 2028 election, where support for a candidate suddenly surges on the prediction market, but no polls or news reports indicate any change—just a mysterious spike. Then CNN starts broadcasting this price 24/7, and voters begin to believe it, thinking this candidate is a guaranteed winner.

The key question is—does this reflect genuine public opinion, or are there behind-the-scenes efforts to manipulate it? There are actually many historical examples of market manipulation. During the 1916 US election, political parties attempted to inflate betting volumes to create momentum. In the 2012 Obama campaign, a trader placed a large order to push Romney’s stock price up by 8 points, which then quickly fell back—classic signs of manipulation.

But the harsh truth is, even if manipulation succeeds, its actual impact on election results is limited. Studies show that voters are not very sensitive to information like "how tight the race is"; the real factors influencing voting are party loyalty and economic conditions. The issue isn’t how much manipulation can sway votes, but how much trust it can destroy. Once people suspect the market is rigged, the panic that "public opinion might be fake" spreads rapidly.

To mitigate this risk, several measures are needed: news outlets should quote only from highly liquid, active markets (which are harder to manipulate), platforms must establish monitoring systems to detect abnormal trades, and transparency should be increased by showing order books. Most importantly, regulators must make it clear—manipulating election prediction markets is illegal.

In essence, prediction markets themselves are not the problem; it’s how they are used. When used properly, they can help understand genuine public opinion. When misused, they become tools for propaganda. This issue is more urgent than polls because polls are questionnaires, while prediction markets involve real money, giving both sides a stronger incentive to cheat.