How US efforts to restrict China from Nvidia’s fastest chips may have backfired

DeepSeek’s rattling of the artificial intelligence market has been several years in the making, with analysts viewing the Chinese company’s shocking advance in the field as an unintended consequence of US trade policy.

Starting under the Biden administration in 2022, the US has imposed export controls to keep China from accessing high-performance semiconductors, primarily from Nvidia. The intention was to keep a technological edge in artificial intelligence and supercomputer-related technologies.

In response, China started creating its own, cheaper chips that came to fruition with DeepSeek, which recently drew attention for its ability to replicate AI functions at a much cheaper cost. Though the US in 2023 tightened policy further in response to the modified chips, that didn’t stop DeepSeek from moving forward.

Commentary over the past several days from industry leaders sent US markets into a tailspin Monday on fears that disruption from DeepSeek could upend a central Wall Street theme that’s relied on Nvidia and related companies to power the market since late 2022 when ChatGPT debuted.

China innovation

“Investors are concerned that rather than impede China’s progress in AI, the US restrictions have engendered innovation that has enabled the development of a model that prioritises efficiency,” JPMorgan Chase analyst Sandeep Deshpande said in a Monday note.

The implications could be significant, but ultimately not all negative for the state of AI in the US and how future developments affect industries. Investors could find buying opportunities in beaten-down companies, while the rise of more cost-efficient technologies opens up the market in other technology, such as notebooks and smartphones.

But more immediately, investors are reexamining the valuation of Nvidia, which trades at 32 times its next 12 months estimated earnings, and whose shares tumbled as much as 18 per cent on Monday.

“What makes Monday’s tech selloff so jarring is that the valuations of many of these AI and tech companies offer no margin of error,” said David Bahnsen, chief investment officer at Bahnsen Group. “Excessive valuation always becomes a problem, eventually, but fundamental news becomes a heightened problem when it is combined with excessive valuation.”

The road ahead

Key to the road ahead will be the extent to which China becomes a bigger player in the AI market and how that plays out not only for Nvidia, but also other companies that are developing their own products and others that are counting on generating the massive amounts of energy required to power the technology.

“Attempts to limit China’s access to US technology, in this case chips and chip equipment, have very likely incentivised China to develop on their own, and become more independent of US capabilities in the process,” Bahnsen said. “China may not only go away as a customer, but may become the #1 competitor as well.”

If nothing else, the latest developments raise more questions about being able to make money from AI, a critical point that has proven largely elusive so far. Investors could be unwilling to approve additional capital expenditures for AI unless monetisation improves.

Also, the US may have to look at its own policies that have sought to manage the spread of AI over industries and around the globe. President Donald Trump has been an outspoken advocate of AI, with close advisor Elon Musk a developer in the field with his xAI product.

“Trump/Musk likely recognise the risk of further restrictions is to force China to innovate faster,” Jefferies analysts said in a note. “Therefore, we think it likely Trump will relax the AI Diffusion policy.”