#预测市场 Seeing Gemini's win rate in the prediction market on Kalshi soar from 30% at the beginning of the year to 86%, this data change is quite interesting. But I want to say that prediction markets are inherently probabilistic games; even a high win rate is just a collective forecast by market participants, not certainty.

This reminds me of an important principle—market sentiment can fluctuate rapidly. At the start of this year, ChatGPT was leading, but the landscape has changed in the blink of an eye. If someone heavily bet based on last year's predictions, they might now be experiencing psychological volatility.

My advice is to maintain a rational watchfulness towards such predictive information. Whether in AI or other fields, the focus is not on guessing the winner correctly, but on: 1. Not letting high-probability numbers hijack your judgment; 2. Even if you are optimistic about a certain direction, control your position size so that a single prediction outcome doesn't threaten your overall asset safety; 3. Give yourself enough time and use long-term holding to hedge against short-term prediction risks.

Steady investors don't make money by betting on predictions; they survive more comfortably through scientific asset allocation and sufficient psychological buffers amid various market changes.