AI stocks at risk as reality bites

The six companies that are yet to report Q2 figures are (expected earnings date):

• Nvidia (23 August)
• Applied Materials (17 August)
• Snowflake (23 August)
• Marvell Technology (24 August)
• Pure Storage (30 August)
• UiPath(6 September)

Are analysts not paying attention in Nvidia?

Nvidia is at the center of the AI hype, but analysts were extremely slow to pay attention earlier this year despite indicators showing demand for chat bots such as ChatGPT was exploding. Revenue estimates for FY25 (ending January 2025) were sticky around the $36-37bn and it was not until Nvidia revealed its outlook for the fiscal year May that analysts paid attention to what was happening. At that point Nvidia shares were already up 109% year-to-date. Post the earnings result and the massive upward revision to revenue growth from Nvidia the share price took off combined with higher revenue estimates from analysts.

With data since June suggesting a significant cooling of interest in AI analysts keep revising their estimates for FY25 revenue now at $59.1bn which is more than a doubling from the FY23 revenue of $27bn. Once again, analysts seem quite slow in updating their estimates to reality.

As we flagged in equity note The Dangerous risk to Nvidia’s exuberant outlook it is quite clear that Nvidia’s growth bonanza is driven by Chinese internet companies playing catchup with Silicon Valley companies ordering massive amount of modified AI chips before the window potentially closes as Chinese companies fear that the Biden administration will prohibit sales of advanced AI chips to China. As we show in our analysis it is the Chinese geographical segment that is driving growth and not the US and this trend will be even more visible when Nvidia report FY24 Q2 (ending 31 July) figures on the 23 August. Because Chinese technology companies are scrambling to catch up with US technology companies there is a real risk that what the market thinks is a sustained high growth rate in AI technology is merely a short-term ‘ketchup effect’ led by Chinese companies. Interestingly enough, the FT recently carried the same message in their recent article China’s internet giants order $5bn of Nvidia chips to power AI ambitions.

In any case, we see accelerating weakness across the stocks in our AI basket and especially Nvidia is key to risk sentiment in AI-related stocks. Because of the significant gains this year in these stocks this group of stocks has a higher fragility because investors may rush for the exit to lock in quick gains made in just seven months if they smell the trend is over. As we have been arguing since June we recommend investors to lower their exposure to AI and semiconductor stocks as expectations have run too high.