The Nasdaq closed 2 per cent up on Tuesday as tech stocks began to recover from Monday’s DeepSeek shock. Investors reacted with panic after the small Chinese firm unveiled its R1 Chatbot, which is thought to be perform comparably to OpenAI’s Chat GPT 4. DeepSeek appeared to implode assumptions which underpin the value of America’s leading AI firms after it claimed its R1 model was created in two months with a budget of just $5 million.

However, markets entered a modest recovery after it became clear that while DeepSeek does appear to have pulled off an efficiency coup, its $5 million price tag likely conceals large procurement expenditures made in contravention of export controls. That understanding has revitalised expectations that large and expensive data centres will still need to be built across the US and the world.

Secondly, while DeepSeek may not be being transparent about the true cost of its R1 model, it is clear the Chinese firm has managed to ‘squeeze out’ more efficiency from its chips, as US AI Tsar David Sacks said on Tuesday. Though this threw investors into panic on Monday, the potential for US rivals to make similar or even more pronounced efficiency savings, and the revelation that DeepSeek was not so far ahead as it first appeared, injected fresh confidence into the markets.

Nigel Green, CEO of the deVere Group, was among the first to tell investors that the emergence of DeepSeek should be seen as an opportunity. Issuing a statement on Monday, the finance chief said:

“DeepSeek’s breakthrough signals a shift toward efficiency in AI, which will redefine both energy and AI markets. The opportunities for investors willing to act now are enormous. This challenges the assumption that AI’s growth is tied to ever-increasing energy consumption.

“While the market is reacting to short-term uncertainty, efficiency-driven AI models will expand adoption into new markets and industries. This means more widespread use, deeper integration, and ultimately, sustained demand for energy solutions.”

With the dust now settling on AI’s ‘Sputnik Moment’ many investors are now coming to the same conclusion, with the Wall Street Journal reporting on Tuesday that: “Monday, stocks acted as if the emergence of DeepSeek’s low-cost approach to AI would be a big problem. Today, the theory in some corners was that cheap AI will ultimately hasten adoption of the technology and benefit the broader industry.”

With questions continuing to swirl around DeepSeek, it appears the Chinese tech firm has made its mark, but not landed a fatal blow to the US AI sector, which now threatens to come back stronger than ever in the face of its new competition.

Microsoft launches investigation into DeepSeek

DeepSeek’s claims that its Chat GPT rival had been developed for just $5 million caused panic and alarm in the markets on Monday. But those claims are now coming under intense scrutiny, with suspicions that the firm may have improperly acquired data and concealed the cost of chips which are banned from being exported to China.

By Wednesday, Microsoft had launched a probe to discover whether a DeepSeek-linked group had been exfiltrating data from OpenAI, a source close to the company told Bloomberg. It comes after security researchers at Microsoft noted that in the Spring of 2024, individuals believed to be connected to DeepSeek were caught exfiltrating large amounts of data from OpenAI – in contravention of its terms of service.

The revelations follow public allegations of the same nature emanating from Trump’s AI Tsar, David Sacks who claimed there was “Substantial evidence” that DeepSeek’s R1 chatbot had improperly distilled information from ChatGPT, in comments to Fox News. Though the investor-turned-government official did concede DeepSeek had made innovations in efficiency – they weren’t as dramatic as had been claimed by the firm on Monday, he said.

Mr Sacks told the outlet that American AI innovators had become complacent and allowed China to catch up with them. He claimed industry leaders became overly focussed on making their learning models politically correct, rather than meaningfully driving the technology forward. He said:

“To be honest, I think that maybe they got a little bit complacent. They didn’t realize how close these Chinese companies were to them. They wasted a lot of time on things like DEI.

“You saw there was like woke AI, there were, you know, the models were basically producing things like black George Washington. And I think that when you’re complacent, you think that there’s not global competition, you can indulge in those sorts of things.”

Crystal van Oosterom, AI venture partner at OpenOcean also waded into the debate, saying it was clear DeepSeek R1 had been trained on its American competition. She told the BBC “DeepSeek has clearly built upon publicly available research from major American and European institutions and companies”.

Stocks stabilise as DeepSeek panic cools

The stock market began to stabilise on Wednesday morning after a Monday rout, with the Nasdaq down 0.7 per cent on Tuesday’s rally. Trading data shows retail investors bought more than $562 million of Nvidia after its record crash with investors looking to buy the dip, with one senior analyst slamming the DeepSeek panic as “overblown.” However, there was caution among investors ahead of a crucial rate decision and Magnificent Seven earnings reports.

European shares broke records on Wednesday after Dutch chipmaker ASML carried the STOXX up to a new high watermark. Shares in the company jumped more than 10 per cent after a strong Q4 report and recovered from a DeepSeek sell-off earlier in the week.

A consensus appeared to be emerging among analysts on Wednesday that Monday’s panicked sell-off was an overaction. Bank of America’s Vivek Arya stocks could be poised to rally, saying in a note that investors should “view the recent selloff as an enhanced buy opportunity.” Separately Jefferies Group said in a note that: “We view DeepSeek’s release as part of an ongoing evolution, not revolution, and think that this market reaction is largely overdone.”

The initial shock of the DeepSeek episode has begun to give way to renewed optimism. As Fortune reported:

“Across Wall Street that seemed to be the prevailing sentiment. Long-term optimism on artificial intelligence remained, resting largely on the view that hundreds of billions in investment into the technology would continue. DeepSeek’s innovative training approach, investors believe, could expand the market for AI to new enterprise customers.

“At the same time, the U.S. will shovel money toward government-funded AI research in an effort to maintain its lead over China. That’s to say nothing of the headwinds DeepSeek could now face as its new-kid-on-the-block hype fades to be replaced by the same level of scrutiny its more established competitors regularly face.”

That optimism could be further buoyed by the prospect of US restrictions on DeepSeek and potentially other Chinese AI firms looking to operate in the US. At his confirmation hearing on Wednesday, Trump’s nominee for Commerce Secretary accused DeepSeek of ‘breaking in and stealing things’ to a panel of senators. In a further boon for US tech prospects, the nominee backed a Biden-era bill which laid down $52 billion to boost US chip production as an “excellent down payment.”

It comes after White House officials were revealed to be looking into potential national security threats posed by DeepSeek. In a White House press conference, press secretary Karoline Leavitt announced the National Security Council was investigating claims of intellectual property theft and added the Trump administration was working to “ensure American AI dominance.”

While at this early stage, it remains unclear what if any action might be taken against DeepSeek by the Trump administration, or if the many claims against the company are born out – never the less there appears to be appetite among senior officials both to bolster US AI and squeeze out competition from abroad. As CBS reported:

“Rep. John Moolenaar, a Michigan Republican who chairs the bipartisan House Select Committee on the Chinese Communist Party, said Tuesday in a statement shared on social media. ‘We must work to swiftly place stronger export controls on technologies critical to DeepSeek’s AI infrastructure.’”

DeepSeek episode could present an opportunity for investors

The sudden emergence of DeepSeek AI put a cat amongst the pigeons. After a bout of heavy turbulence prompted by the unveiling of the R1 Chatbot, investors have adopted a more measured posture – seeing the potential in AI efficiencies and seeing through some of DeepSeek’s initial claims.

DeepSeek’s assertion that it developed a ChatGPT-4 rival in just two months with a budget of $5 million sent shockwaves through the tech industry. However, doubts are now emerging over the veracity of these claims, with suggestions that the company may be concealing the true cost and the method by which the chatbot was trained. Statements from public officials might indicate that DeepSeek may have improperly acquired data and utilised chips subject to export controls.

In response to these revelations, major industry players have initiated probes to assess the extent of DeepSeek’s actions. Microsoft and OpenAI are investigating whether DeepSeek illicitly obtained data from ChatGPT to train its own models. These investigations underscore the importance of maintaining ethical standards and protecting intellectual property within the AI sector.

The U.S. government, in line with its America First posture, has been quick to respond to developments. The White House is evaluating the national security implications of DeepSeek’s AI applications, reflecting concerns over data privacy and the potential misuse of AI technologies. This scrutiny aligns with broader efforts to ensure that AI advancements adhere to ethical guidelines and do not compromise national interests.

By Wednesday markets began to reflect investor sentiment that while DeepSeek’s approach highlights the potential for more efficient AI development, it did not necessarily undermine the foundational strengths of established firms. The emphasis on efficiency and cost-effectiveness may drive innovation across the sector, prompting companies to explore new methodologies and optimise resource utilisation, allowing them to get more for less.

However, while fundamentals remain strong, the situation remains uncertain and there could be more volatility in the year ahead. The CEO of the deVere Group, Nigel Green has repeatedly warned investors that they should take a cautious approach to investing this year.

Mr Green, the head of one of the world’s leading financial institutions, believes the looming spectre of inflation, geopolitical uncertainty and emerging technologies means investors should now more than ever ensure they have a diversified and resilient portfolio. A correction could be on the horizon, and whether that comes in the shape of a DeepSeek-style tech upstart, unexpected inflationary figures or a bubble bursting, investors should carefully evaluate their positions. Speaking earlier this month Mr Green said:

“The US economy is going fast. So why would the stock market crash? If interest rates stay high the market has priced in interest rates coming down. People are running out of money. In other words, whilst the job market is strong people have been overspending. Consumer defaults are at their highest rate in ten years.

“We have a situation where inflation is high and The Fed can’t cut interest rates for at least six months. Potentially they may not even cut in 2025, I am predicting a single cut.”

However Mr Green said he believed AI stocks, the energy sector and the financial industry would continue to do well, saying “Trump will deregulate and cut red tape, so those companies potentially can continue to do well.”

Investors appear to be coming away from the DeepSeek episode not with despair for the future of American AI, but renewed hope that the likes of Open AI can become more efficient and roll out the many productivity-boosting potential of AI across industry at pace. Meanwhile, a strong reaction by US officials indicates that the prospective value of any Chinese competition in the AI sector could be stunted by US regulatory regimes, with officials citing concerns over intellectual property theft and cybersecurity.