Nvidia lost $178 billion in a single day, and smart money predicts its market value will be overtaken by Apple.

According to PolyBeats monitoring, over the weekend, heads of multiple frontier AI labs collectively supported slowing down model capability iteration, causing Nvidia to fall 3.36% on Monday in a single day, evaporating about $178 billion. Based on market cap data around the close of U.S. stocks on September 14, Nvidia is still about $222 billion higher than Apple. If Nvidia's market cap does not move, Apple needs to gain another roughly 4.5% in relative terms to overtake it.

Yesterday, Hauchn put in $2.3k on "Apple will become the world's most valuable company at this month's close," and the current "Yes" probability is 14.9%.

Hauchn's best-related category on this market is Business, with a category net profit of $404k. In this category, their win rate across 181 settled trades is 70/181 (39%), among which there are 7 trades with a buy price below $0.8 and a sell price above $0.95. Within a similar cost price range ($0.051-$0.2), their historical median amount invested is $1.4k. This account's prediction-type trades (held to settlement) have a win rate of only 4.3% (4/93).

On Apple's side, Morgan Stanley maintains a bullish view, believing that the iPhone Duo, A20 chip, and Siri updates are expected to push Apple's market cap to about $5.25T; however, KeyBanc maintains an underweight view, with a corresponding target market cap of only about $3.65T, implying Apple's market cap could shrink by about $1.25T from current levels, with its concern being that higher prices for new models will suppress demand. Both are medium- to long-term target prices, not direct forecasts for the end of September.

On Nvidia's side, Baird recently still maintained an outperform rating, with the core logic being demand for inference compute and enterprise AI infrastructure; however, Monday's decline shows that what the market is paying more attention to this month is not long-term target prices, but whether "slowing down" will evolve from CEO statements into actual reductions in training, delayed model releases, or lowered data center capital expenditures.