Semiconductors Slide 8% in July as Chinese AI Model Rattles Chip Investors

The semiconductor sector took another hard hit on Friday, with the VanEck Semiconductor ETF (SMH) falling more than 4% in the session and extending a brutal July selloff that has now erased more than 8% from the ETF’s value. This marks the third weekly decline in four weeks for the SMH. The catalyst was a Chinese startup. Moonshot AI unveiled a new artificial intelligence model this week that the company claims performs on par with the leading offerings from OpenAI and Anthropic — at a fraction of the infrastructure cost. Applied Materials, Lam Research, Intel, KLA Corp., and Arm Holdings each fell roughly 4%, while Nvidia and Micron Technology dropped more than 2%.

The market reaction reveals a growing fear: if Chinese startups can build competitive AI models using cheaper hardware and open-source techniques, the multi-hundred-billion-dollar buildout of U.S. AI data centers could be overbuilt. “We are seeing signs of fatigue, with end-user demand for AI becoming more price sensitive and the market starting to penalize companies that are ramping spending too aggressively,” said Angelo Kourkafas, senior investment strategist at Edward Jones. The Moonshot AI news echoes the DeepSeek shock from earlier this year that sent chip stocks tumbling, when the Chinese model demonstrated that U.S. companies were possibly over-investing in compute infrastructure. This latest event reinforces the same narrative — that AI efficiency gains are outpacing raw compute demand, putting pressure on semiconductor revenue forecasts. Adding to the pressure, New York Governor Kathy Hochul signed an executive order this week barring new large-scale data centers for up to one year, with at least 15 other states considering similar restrictions.

  • Special: THE STARLINK OF ENERGY. This Stock May Benefit From a Major Gov't Catalyst
  • For retail investors, the chip sector now sits at a decision point. The structural AI demand thesis hasn’t been disproven — hyperscalers like Microsoft, Meta, and Google are still spending aggressively on data center capacity. But the short-term narrative has shifted toward “how much compute is really needed?” which puts pressure on valuation multiples. The S&P 500 fell 0.9% Friday while the Nasdaq dropped 1.2%, with chips the biggest drag. Investors with existing chip exposure through ETFs like SMH or SOXX likely shouldn’t panic-sell — but the easy money from the 2023-2024 AI infrastructure trade has been made. Those looking to add might wait for the Moonshot AI news to be fully digested, or consider rotating toward chip equipment names with long order backlogs that are less sensitive to model efficiency improvements.