When everyone is looking at something, it changes
In quantum physics, the observer effect is the theory that simply observing a situation or phenomenon necessarily changes that phenomenon.
Economics has something similar, called Goodhart's law. It's named after British economist Charles Goodhart who observed that when a measure becomes a target, it ceases to be a good measure.
The issue is that once everyone is watching a measure or a target and knows what it means, then the observers anticipate the effects and adjust for the outcome, thus changing the effect.
The idea is prescient because of the market's intense focus on the yield curve inversion. When longer-dated yields fall below shorter-dated yields, it's a classic recessionary signal. Intuitively it makes sense because it signals a situation where people would rather have long-term safety than earn more yield in the near-term.
Its effectiveness as a predictor of recessions is undoubtedly excellent but whether the curve is inverted is always a matter of interpretation. Surely 10-year yields falling below 3-month bill yields is an inversion but is 5-year yields falling below 2-year yields as strong of a signal?
More importantly, does Goodhart's law apply? If all market participants believe it's a sure sign of a recession, then they will surely behave differently than they would otherwise. Similarly, central bankers are also watching the yield curve more closely than ever and as the curve approaches inversion, they may react.
Taken to a further extreme, you may have central banker who actively target the entire yield curve in the belief that so long as the curve is steep, a recession can't occur.
The idea of Goodhart's law is that any statistical relationship will break down when used for policy purposes.
This is frequently seen in regulation and even in the workplace. Whenever targets are set, people find ways to game and abuse them, and so do broad markets and the economy.