Trump calls China’s DeepSeek AI app a ‘wake-up call’

Trump calls China’s DeepSeek AI app a ‘wake-up call’ after tech stocks slide

Silicon Valley and Washington leaders said the app shows China can challenge the U.S. The Nasdaq lost 3 percent and chipmaker Nvidia shed $589 billion in market capitalization.

By Eva Dou, Aaron Gregg, Cat Zakrzewski, Nitasha Tiku and Shannon Najmabadi WP 28-01-2025

The sudden popularity of a Chinese artificial intelligence app called DeepSeek pummeled tech stocks and captivated Silicon Valley on Monday, prompting debate in political and tech industry circles about how the United States can maintain its lead in AI.

Get a curated selection of 10 of our best stories in your inbox every weekend.

President Donald Trump said late Monday that DeepSeek’s release “should be a wake-up call for our industries that we need to be laser focused on competing to win.”

Trump separately said he planned to place new tariffs on computer chips produced overseas, a policy that could create further challenges for the U.S. tech industry, which depends heavily on chip manufacturing in Asia.

The DeepSeek app rocketed to the top of the downloads chart in the Apple store over the weekend and remained there Monday after its release last week by a Chinese start-up of the same name founded in 2023. The app offers similar functionality to OpenAI’s popular ChatGPT chatbot, answering questions and generating text in response to a user’s queries.

Several tech companies that have banked on a surge of AI interest sold off Monday, with U.S. chipmaker Nvidia down almost 17 percent, losing $589 billion in market capitalization. The tech-focused Nasdaq composite index finished the day down 3 percent.

DeepSeek’s explosive debut also escalated concerns about China’s ability to challenge the U.S. lead in advanced artificial intelligence. Both nations have positioned prowess in AI technology as central to their future economic and military power.

The Biden administration introduced several measures intended to restrict Chinese AI development, including sweeping new export controls this month to limit the country’s access to powerful chips known as GPUs, which underpin advanced AI projects. Those latest controls were in part aimed at blocking Chinese efforts to circumvent previous export controls.

Victoria LaCivita, a spokeswoman for the White House Office of Science and Technology, said former president Joe Biden’s policies had failed to limit access to American technology and created an opportunity for China and other foreign adversaries to make gains in AI development. David Sacks, President Donald Trump’s AI and crypto czar, said in a post on X that DeepSeek “shows that the AI race will be very competitive.”

The Trump administration has shared few specifics about its own approach to AI policy. The president last week rescinded a sweeping executive order on AI signed by Biden in 2023 and signed an executive order of his own directing agencies to rescind all actions taken under the Biden order “that are inconsistent with enhancing America’s leadership in AI.”

Trump in his remarks late Monday in Doral, Florida, said that the U.S. had the “greatest scientists in the world” and that American companies could benefit from building on DeepSeek’s ideas.

He also criticized the billions of dollars of grants provided to encourage U.S. semiconductor manufacturing under the Biden administration, although the 2022 Chips Act that provided the funding received bipartisan support at the time. Trump said that tariffs he planned to place on imports of foreign-made chips would be more effective, but did not specify details of that policy or when it might take effect.

DeepSeek was founded by Liang Wenfeng, co-founder of the hedge fund High-Flyer, and says it has developed ways to create the AI models needed to power chatbots and other tools more cheaply. It claims to need fewer and less-advanced chips to develop AI software that can compete with technology from U.S. rivals.

Analysts said the Monday sell-off underscores anxieties about whether the massive recent spending by U.S. firms on specialized chips, data centers and related power infrastructure to power AI projects is justified.

Nvidia in particular has exploded in value in recent years because it dominates the market for the GPU chips at the center of the global AI race. It’s now one of three companies with a market capitalization above $3 trillion.

Leading tech firms have spent billions building out artificial intelligence technology for sale to large businesses. Meta chief executive Mark Zuckerberg said in a Facebook post last week that his company plans to invest between $60 billion and $65 billion on AI and build a massive data center. OpenAI, Oracle and SoftBank are leading the Stargate venture announced with Trump last week that seeks to spend up to $500 billion building out data centers to support AI projects.

Monday’s sell-off also shaved 1.46 percent off the broader S&P 500 index. The Dow Jones Industrial Average, which contains fewer tech stocks, gained 0.6 percent. Wall Street’s fear gauge, the CBOE Volatility Index, climbed almost 22 percent.

Some analysts suggested that Wall Street’s reaction may be a premature reflection of fear in the markets, noting that U.S. companies have dominated artificial intelligence innovation so far. WedBush senior analyst Dan Ives, who has been bullish on AI, called Monday’s rout a “golden buying opportunity” despite the reaction in U.S. markets.⍐

China’s AI startup sending shockwaves through global tech

What’s DeepSeek, China’s AI startup sending shockwaves through global tech?

Chip giant Nvidia shed nearly $600bn in market value after Chinese AI model cast doubt on supremacy of US tech firms.By John Power Published On 28 Jan 2025 AJ 

DeepSeek, a little-known Chinese startup, has sent shockwaves through the global tech sector with the release of an artificial intelligence (AI) model whose capabilities rival the creations of Google and OpenAI.DeepSeek-R1’s creator says its model was developed using less advanced, and fewer, computer chips than those employed by tech giants in the United States.In a research paper released last week, the model’s development team said they had spent less than $6m on computing power to train the model – a fraction of the multibillion-dollar AI budgets enjoyed by US tech giants such as OpenAI, Alphabet and Meta.Marc Andreessen, one of the most influential tech venture capitalists in Silicon Valley, hailed the release of the model as “AI’s Sputnik moment”.The sudden emergence of a small Chinese startup capable of rivalling Silicon Valley’s top players has challenged assumptions about US dominance in AI and raised fears that the sky-high market valuations of companies such as Nvidia, Alphabet and Meta may be detached from reality.On Monday, Nvidia, which holds a near-monopoly on producing the semiconductors that power generative AI, lost nearly $600bn in market capitalisation after its shares plummeted 17 percent.US President Donald Trump, who last week announced the launch of a $500bn AI initiative led by OpenAI, Texas-based Oracle and Japan’s SoftBank, said DeepSeek should serve as a “wake-up call” on the need for US industry to be “laser-focused on competing to win”.

What is DeepSeek?

DeepSeek, which is based in Hangzhou, was founded in late 2023 by Liang Wenfeng, a serial entrepreneur who also runs the hedge fund High-Flyer.

Though little known outside China, Liang has an extensive history of combining burgeoning technologies and investing.

In 2013, he co-founded Hangzhou Jacobi Investment Management, an investment firm that employed AI to implement trading strategies, along with a co-alumnus of Zhejiang University, according to Chinese media outlet Sina Finance.

Liang went on to establish two more firms focused on computer-directed investment – Hangzhou Huanfang Technology Co and Ningbo Huanfang Quantitative Investment Management Partnership – in 2015 and 2016, respectively.

In an interview with Chinese media outlet Waves in 2023, Liang dismissed the suggestion that it was too late for startups to get involved in AI or that it should be considered prohibitively costly.

“Reproduction alone is relatively cheap — based on public papers and open-source code, minimal times of training, or even fine-tuning, suffices. Research, however, involves extensive experiments, comparisons, and higher computational and talent demands,” Liang said, according to a translation of his comments published by the ChinaTalk Substack.

Liang said his interest in AI was driven primarily by “curiosity”.

“From a broader perspective, we want to validate certain hypotheses. For example, we hypothesise that the essence of human intelligence might be language, and human thought could essentially be a linguistic process,” he said, according to the transcript.

“What you think of as ‘thinking’ might actually be your brain weaving language. This suggests that human-like AGI could potentially emerge from large language models,” he added, referring to artificial general intelligence (AGI), a type of AI that attempts to imitate the cognitive abilities of the human mind.

DeepSeek did not immediately respond to a request for comment.

On Monday, Gregory Zuckerman, a journalist with The Wall Street Journal, said he had learned that Liang, who he had not heard of previously, wrote the preface for the Chinese edition of a book he authored about the late American hedge fund manager Jim Simons.

“Simons left a deep impact, apparently,” Zuckerman wrote in a column, describing how Liang praised his book as a tome that “unravels many previously unresolved mysteries and brings us a wealth of experiences to learn from”.

“Even my mother didn’t get that much out of the book,” Zuckerman wrote. 

 

DeepSeek’s research paper suggests that either the most advanced chips are not needed to create high-performing AI models or that Chinese firms can still source chips in sufficient quantities – or a combination of both.

California-based Nvidia’s H800 chips, which were designed to comply with US export controls, were freely exported to China until October 2023, when the administration of then-President Joe Biden added them to its list of restricted items.

In his 2023 interview with Waves, Liang said his company had stockpiled 10,000 Nvidia A100 GPUs before they were banned for export. GPUs, or graphics processing units, are electronic circuits used to speed up graphics and image processing on computing devices.

Tanishq Abraham, former research director at Stability AI, said he was not surprised by China’s level of progress in AI given the rollout of various models by Chinese firms such as Alibaba and Baichuan.

“While there have been restrictions on China’s ability to obtain GPUs, China still has managed to innovate and squeeze performance out of whatever they have,” Abraham told Al Jazeera.

“I think it is a lesson to US companies that there is still a lot of performance they can squeeze out of.”

Tara Javidi, co-director of the Center for Machine Intelligence, Computing and Security at the University of California San Diego, said DeepSeek made her excited about the “rapid progress” taking place in AI development worldwide.

“My only hope is that the attention given to this announcement will foster greater intellectual interest in the topic, further expand the talent pool, and, last but not least, increase both private and public investment in AI research in the US,” Javidi told Al Jazeera.

Meanwhile, investors’ confidence in the US tech scene has taken a hit – at least in the short term.

Apart from Nvidia’s dramatic slide, Google parent Alphabet and Microsoft on Monday saw their stock prices fall 4.03 percent and 2.14 percent, respectively, though Apple and Amazon finished higher.

“If DeepSeek’s cost numbers are real, then now pretty much any large organisation in any company can build on and host it,” Tim Miller, a professor specialising in AI at the University of Queensland, told Al Jazeera.

“So, in this sense, the game has changed completely because there is a new ‘rule’ that anyone can play.”

Does this mean China is winning the AI race?

Not necessarily.

While tech analysts broadly agree that DeepSeek-R1 performs at a similar level to ChatGPT – or even better for certain tasks – the field is moving fast.

OpenAI CEO Sam Altman said earlier this month that the company would release its latest reasoning AI model, o3 mini, within weeks after considering user feedback.

On Monday, Altman acknowledged that DeepSeek-R1 was “impressive” while defending his company’s focus on greater computing power.

“We will obviously deliver much better models and also it’s legit invigorating to have a new competitor! We will pull up some releases,” Altman said on X.

“But mostly we are excited to continue to execute on our research roadmap and believe more compute is more important now than ever before to succeed at our mission.”

“Most entrepreneurs had completely missed the opportunity that generative AI represented, and felt very humbled,” Ma told Al Jazeera.

“It’s clear that they have been hard at work since. I think what this past weekend shows us is how seriously they self-reflected and took the challenge to ‘catch up’ to Silicon Valley. I think that for the US to retain its lead, Washington should focus on boosting Silicon Valley instead of on suppressing China.”

Abraham, the former research director at Stability AI, said perceptions may also be skewed by the fact that, unlike DeepSeek, companies such as OpenAI have not made their most advanced models freely available to the public.

“DeepSeek made its best model available for free to use. On the other hand, OpenAI’s best model is not free,” he said.

“So most people who use ChatGPT for free are shocked by DeepSeek and believe there is a huge jump in capabilities when OpenAI has had a similar performing model paywalled for a few months already. This pay-walling of frontier AI models leads to people not truly grasping the progress and capabilities of AI.”

Miller, the University of Queensland professor, said DeepSeek’s advances and other recent developments suggest that China is at least “up there” with the US in AI.

“I made somewhat of a throwaway prediction late last year that the next scientific breakthrough in AI could come from a small player such as an individual university researcher who doesn’t have access to much computing power – they would need to be smarter to compete,” he said.

“DeepSeek’s apparent progress is almost an example of this: by not having enough computational power to build models as large as ChatGPT, they had to be smart. Necessity is the mother of invention.”

Source: Al Jazeera 28-01-2025 

 

DeepSeek launches new AI model

 DeepSeek launches new AI model as Trump cautions of ‘wake-up call’ to US industry

By Ma TongPublished: Jan 28, 2025

Chinese artificial intelligence (AI) startup DeepSeek on Tuesday launched a new open-source multimodal model, following the buzz generated by its cost-effective open-source reasoning model, DeepSeek-R1, which competes with rivals like OpenAI but at a significantly lower cost, sending ripples through the US stock market the day before.

According to information on the AI community platform Hugging Face on Tuesday, DeepSeek has released the open-source multimodal AI model Janus-Pro, an upgraded version of its earlier Janus model, which significantly enhances multimodal understanding and visual generation capabilities.

Its Janus-Pro-7B AI model outperformed OpenAI's DALL-E 3 and Stability AI's Stable Diffusion in a leaderboard ranking for image generation using text prompts, Reuters reported on Tuesday, according to a DeepSeek's technical report the Global Times read on Github, a proprietary developer platform. 

The latest development came after earlier in January the company released the latest open-source model DeepSeek-R1, which has achieved an important technological breakthrough - using pure deep learning methods to allow AI to spontaneously emerge with reasoning capabilities.

As a rapidly growing competitor to leading AI tools like OpenAI's ChatGPT, Google's Gemini, and others, the Chinese AI startup has garnered runaway attention in recent weeks.

US President Donald Trump said on Monday that Chinese startup DeepSeek's technology should act as spur for American companies and said it was good that companies in China have come up with a cheaper, faster method of artificial intelligence, Reuters reported Tuesday. 

"I've been reading about China and some of the companies in China, one in particular coming up with a faster method of AI and much less expensive method, and that's good because you don't have to spend as much money. I view that as a positive, as an asset," Trump said, according to Reuters. 

"The release of DeepSeek, AI from a Chinese company should be a wakeup call for our industries that we need to be laser-focused on competing to win," Trump said in Florida, according to the report. 

The remarks also came as the news around DeepSeek sent shockwaves through the AI industry, with Nvidia bearing the brunt of the sell-off. The chipmaker, a linchpin of the AI supply chain, lost over $500 billion in market value, plummeting 16.86 percent in a single day. Other major tech players, including Alphabet and Microsoft, also declined, though Meta managed to trade in positive territory, Xinhua reported. 

Nvidia issued a statement on Monday after its shares tumbled, noting DeepSeek's advances show the usefulness of its chips for the Chinese market and that more of its chips will be needed in the future to meet demand for DeepSeek's services.

"DeepSeek's work illustrates how new models can be created using that technique, leveraging widely-available models and compute that is fully export control compliant," Nvidia said in its statement.

OpenAI Chief Executive Officer Sam Altman welcomed the debut of DeepSeek's R1 model in a post on X late on Monday. "Deepseek's r1 is an impressive model, particularly around what they're able to deliver for the price. we will obviously deliver much better models and also it's legit invigorating to have a new competitor! we will pull up some releases," Altman said.

The success of DeepSeek showed that the Biden administration's four-year crackdown on China's AI and computing power has not only failed but has also spurred the country to forge a unique path for AI development, achieving significant progress in autonomous AI development, Ma Jihua, a veteran telecom industry observer, told the Global Times on Tuesday.

"While the global AI community has been focused on increasing computing power, China has been pioneering a path through algorithm optimization, opening up a new approach that is cost-effective and equally efficient. This development holds significant importance for the global AI landscape," Ma noted.

"However, with the emergence of DeepSeek and the rapid advancement of China's AI industry, there is now greater potential for complementary cooperation between China and the US. Both countries can leverage their unique strengths, making collaboration more promising than ever before," Ma said. 

Industry observers reached by the Global Times previously said while China and the US, as the two leading powers in the global AI field, compete in the AI industry, there is also significant room for cooperation, especially in AI governance.⍐

OpenAI asks Indian court to throw out book publishers challenge in copyright battle

OpenAI asks Indian court to throw out book publishers challenge in copyright battle

By Arpan Chaturvedi and Aditya Kalra January 28, 2025

• Summary
• 
  

• OpenAI filing says publishers body's case should be dismissed
• OpenAI denies using original literary work in training its large language model

NEW DELHI, Jan 28 (Reuters) - OpenAI has asked an Indian court to quash a plea by a group representing Indian and global book publishers that accuse it of copyright breaches, arguing its ChatGPT service only disseminates public information, legal papers show.

The case, which began with legal action last year by local news agency ANI, will be heard in New Delhi on Tuesday. It has the potential to shape the legal framework for artificial intelligence in India - OpenAI's second-largest market by number of users.

In recent weeks, book publishers and almost a dozen digital media outlets, including those owned by billionaires Gautam Adani and Mukesh Ambani, have joined the case to challenge the AI giant.

The Federation of Indian Publishers, which represents many Indian firms and likes of Bloomsbury (BLPU.L), opens new tab and Penguin Random House, has argued ChatGPT produces book summaries and extracts from unlicensed online copies, hurting their business.

OpenAI countered that the information was drawn from platforms like Wikipedia or abstracts, summaries, tables of content made publicly available on the websites of the publishers in question, according to a Jan. 26 non-public court filing seen by Reuters.

"Web-crawlers are designed to only access publically available data," OpenAI said in its 21-page response to the book publishers's argument.

The book publishers have "entirely failed to demonstrate even a single instance" that OpenAI services are trained on "original literary work," it said.

Pranav Gupta, secretary of the federation, told Reuters that most book-related content being shown by ChatGPT was scraped from websites that have licensing arrangements with book publishers.

OpenAI maintains it only uses publicly available data in a manner protected by fair use principles. Asked for comment on Tuesday, it referred Reuters to its earlier statements and the court filing challenging the book publishers.

OpenAI has also said, in its initial response to the ANI case, that Indian judges have no jurisdiction to hear a case against it as its servers are located abroad.

The case is one of many that is being heard globally in which authors, news organisations and musicians have accused technology firms of using their copyrighted work to train AI services without permission or license.⍐

China's DeepSeek AI assistant surges in popularity

 Tech selloff subsides, DeepSeek triggers AI rethink

By Ankur Banerjee and Rae Wee January 28, 2025

• Summary
• 
  

• China's DeepSeek AI assistant surges in popularity
• Investors ponder AI race between China, U.S.
• Nvidia shares sink 17%, wiping $593 bln off market value
• Japanese technology shares slide for second day

The deepseek logo, a keyboard, and robot hands are seen in this illustration taken January 27, 2025. REUTERS/Dado Ruvic/Illustration

LONDON/SINGAPORE, Jan 28 (Reuters) - Global technology shares recovered some poise on Tuesday, but remained vulnerable after a rout sparked by the emergence of a low-cost Chinese artificial intelligence model made investors question the sky-high valuation and dominance of AI bellwethers.

Shares of chipmaker Nvidia (NVDA.O), opens new tab, the poster child of the AI boom in recent years, fell 17% on Monday, wiping $593 billion from its market value - a record one-day loss for any company - and dragging U.S. stocks lower.

By Tuesday, Nvidia shares were up nearly 6% in Frankfurt, while those in Oracle (ORCL.N), opens new tab rose 3.4% and AI data analytics company Palantir (PLTR.O), opens new tab rose 2.97%.

It all stemmed from a free AI assistant launched by Chinese startup DeepSeek last week that the firm said uses less data at a fraction of the cost of services available currently. That garnered attention worldwide, although scepticism lingers.

OpenAI CEO Sam Altman called it an "impressive model".

"We will obviously deliver much better models and also it's legit invigorating to have a new competitor!," Altman, the head of the AI firm behind ChatGPT, said, opens new tab in a social media post.

The launch and increasing popularity of DeepSeek spurred investors to dump tech stocks globally, with ripples felt from Tokyo to Amsterdam to Silicon Valley.

Markets in tech-heavy South Korea and Taiwan are closed for the next few days for Lunar New Year. Mainland China is closed until Feb. 4, leaving the spotlight firmly on Japanese firms.

On Tuesday, chip-testing equipment maker Advantest (6857.T), opens new tab, a supplier to Nvidia lost 10% after diving nearly 9% on Monday. Chip-making equipment maker Tokyo Electron (8035.T), opens new tab and technology start-up investor SoftBank Group (9984.T), opens new tab slid 5%.

"It's clearly a sell first, ask questions later approach, and we've actually seen that kind of move in the past in Japan," said Kei Okamura, a portfolio manager at Neuberger Berman, referring to a global market meltdown in August headlined by Japan's Nikkei.

In Europe on Tuesday, shares in Dutch semiconductor company ASML (ASML.AS), opens new tab, which closed down 7.1% on Monday, opened up 0.9%, while shares in BE Semiconductor (BESI.AS), opens new tab rose 1.2%.

Over in the U.S., Broadcom (AVGO.O), opens new tab finished down 17.4%, while ChatGPT backer Microsoft (MSFT.O), opens new tab fell 2.1% and Google parent Alphabet (GOOGL.O), opens new tab closed down 4.2%. The Philadelphia semiconductor index (.SOX), opens new tab tumbled 9.2%, for its deepest percentage drop since March 2020.

NO MARGIN OF ERROR

The selloff has brought into the spotlight the crowded positioning among investors and the billions of dollars U.S. tech giants are pouring in to develop AI capabilities, as well as the extremely high valuation of some of these firms.

"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 The Bahnsen Group.

"The excessive weighting these tech stocks have in many investor portfolios and the high concentration these tech stocks have in the market indices was a significant and under-appreciated risk issue."

The hype around AI has powered a huge flow of capital into equities, inflating valuations and lifting stock markets to record highs, leading to an increase of around $10 trillion in the market value of "Magnificent Seven" companies since ChatGPT kicked off the AI boom in November 2022.

It is not just the chipmakers and tech companies but companies focused on data centres also taking a hit, with Malaysia's utility conglomerate YTL Power (YTLP.KL), opens new tab down 9% on Tuesday, its third session of steep loss.

"We're still, like many investors, gathering information," said Neuberger Berman's Okamura, noting that a lot of investors are scrambling to gather more information and decide their next move.

"I think we’re going to see many more of these (developments) going forward. And we’ve seen technological advancements like this that have had implications for cost spend."

Investor focus will be on the flurry of tech earnings this week, with executives likely keen to calm frayed nerves and ease concerns about capital spending.

AI RACE

Little is known about the Hangzhou startup behind DeepSeek, whose controlling shareholder is Liang Wenfeng, co-founder of quantitative hedge fund High-Flyer, records showed.

Its researchers wrote in a paper last month that its DeepSeek-V3 model, launched on Jan. 10, used Nvidia's lower-capability H800 chips for training, at a cost of less than $6 million.

The launch and the popularity of the app contrasts with the lacklustre reception that met the Chinese ChatGPT equivalent made by search engine giant Baidu (9888.HK), opens new tab, which exposed the gap in AI capabilities between U.S. and Chinese firms.

An area chart titled "The Magnificent Seven’s slice of the S&P 500" that tracks the metric.

The quality and cost efficiency of DeepSeek's models have flipped this narrative on its head and prompted a warning from U.S. President Donald Trump, who called it "a wakeup call for our industries."

Japan's digital minister Masaaki Taira said DeepSeek's emergence had upended conventional wisdom that Chinese AI was years behind.

"It's been said that Chinese generative AI might be about five years behind, but that turned out to be wrong and it seems to be on a fairly good track," Taira said, adding that Japan was taking a closer look into suggestions that Chinese AI may be more cost effective.

An area chart titled "The Magnificent Seven’s slice of the S&P 500" that tracks the metric.

The quality and cost efficiency of DeepSeek's models have flipped this narrative on its head and prompted a warning from U.S. President Donald Trump, who called it "a wakeup call for our industries."

Japan's digital minister Masaaki Taira said DeepSeek's emergence had upended conventional wisdom that Chinese AI was years behind.

"It's been said that Chinese generative AI might be about five years behind, but that turned out to be wrong and it seems to be on a fairly good track," Taira said, adding that Japan was taking a closer look into suggestions that Chinese AI may be more cost effective.⍐

China's Deep Seek prompts global sell-off in AI-linked stocks

By Sinéad Carew, Amanda Cooper and Ankur Banerjee

January 27, 2025

• Summary
• 
  

• Tech stocks sink; Nvidia drops sharply
• China's Deep Seek AI assistance surges in popularity
• Nervous investors seek safe havens, dollar falls

LONDON/SINGAPORE, Jan 27 (Reuters) - Investors made a quick exit from a host of technology stocks from Tokyo to New York on Monday as the emergence of a low-cost Chinese artificial intelligence model challenged the dominance of current AI leaders such as Nvidia (NVDA.O), opens new tab.

Raising questions about the level of investment needed for AI, startup DeepSeek launched a free AI assistant last week that it says uses less data at a fraction of the cost of incumbent services.

And by Monday DeepSeek's assistant had overtaken U.S. rival ChatGPT in downloads from Apple's app store.

The news led Nasdaq (.IXIC), opens new tab to open down more than 3%. The tech-heavy index has since pared losses but was still down 2.9% at midday, with leading AI chipmaker Nvidia (NVDA.O), opens new tab its biggest drag with a loss of more than 15%.

The Nasdaq's next biggest drag was chip maker Broadcom Inc (AVGO.O), opens new tab, down 15%, followed by Microsoft (MSFT.O), opens new tab, off 3.7%, and Google parent Alphabet (GOOGL.O), opens new tab, which fell 2.7%.

The Philadelphia semiconductor index (.SOX), opens new tab tumbled 7.9%, eyeing its biggest percentage drop since March 2020.

Their declines followed a sell-off that started in Asia, with Japan's SoftBank Group (9984.T), opens new tab finishing down 8.3%, and moved through Europe where ASML (ASML.AS), opens new tab fell 7.6%.

"If it’s true that DeepSeek is the proverbial 'better mousetrap,' that could disrupt the entire AI narrative that has helped drive the markets over the last two years," said Brian Jacobsen, chief economist at Annex Wealth Management in Menomonee Falls, Wisconsin,

"It could mean less demand for chips, less need for a massive build-out of power production to fuel the models, and less need for large-scale data centers. However, it could also mean that AI becomes more accessible and help kickstart the development of a wide array of useful applications," Jacobsen said.

The hype around AI has powered a huge inflow of capital into the equity markets in the last 18 months, as investors bought into the technology, which inflated company valuations and lifted stock markets to new highs.

On Wednesday U.S. AI-related stocks rallied after President Donald Trump announced a private-sector plan for what he said would be a $500 billion investment in AI infrastructure through a joint venture known as Stargate.

Since then SoftBank has announced a $19 billion commitment to help fund the Stargate joint venture with companies including OpenAI and Oracle , whose shares were down 7.8% on Monday.

U.S.-traded shares in ASML, which counts Taiwan's TSMC (2330.TW), opens new tab, Intel (INTC.O), opens new tab and Samsung (005930.KS), opens new tab as its customers, dropped more than 6%. In Europe energy technology company Siemens Energy (ENR1n.DE), opens new tab lost 19%.

Little is known about the small Hangzhou startup behind DeepSeek. Its researchers wrote in a paper last month the DeepSeek-V3 model, launched on Jan. 10, used Nvidia's H800 chips for training, spending less than $6 million - the figure referenced by Pictet's Withaar.

H800 chips are not top of the line. Initially developed as a reduced-capability product to get around curbs on sales to China, they were subsequently banned by U.S. sanctions.

'SPUTNIK MOMENT'

Marc Andreessen, the Silicon Valley venture capitalist, said in a post on X on Sunday that DeepSeek's R1 model was AI's "Sputnik moment," referencing the former Soviet Union's launch of a satellite that marked the start of the space race in the late 1950s.

"DeepSeek R1 is one of the most amazing and impressive breakthroughs I've ever seen — and as open source, a profound gift to the world," he said in a separate post.

Given the volatility in equities, investors sought out safe-havens such as U.S. Treasuries. This pushed 10-year yields down to 4.55% while currencies like Japan's yen and the Swiss franc rallied against the dollar.

Big Tech has ramped up spending on developing AI capabilities and optimism over the possible returns has driven stock valuations sky-high.

Shares in Nvidia are down more than 7% for the year to date but the stock rose 171% in 2024 and 239% in 2023 to trades at 56 times the value of its earnings.

In comparison the Nasdaq (.IXIC), opens new tab trades at a multiple of 16 to the value of its constituents' earnings, LSEG data showed.

Among other stocks, Vertiv Holdings (VRT.N), opens new tab, which builds data center infrastructure, slumped 27%.

Power utilities, expected to benefit from a surge in demand from power-hungry data centers needed to develop AI technology, also tumbled. Vistra (VST.N), opens new tab shares were down 24.4% while Constellation Energy (CEG.O), opens new tab shares were off more than 18% and NRG Energy (NRG.N), opens new tab fell 13%.

A line chart of the relative performance in percentage terms since May 2023 of the big US chipmakers and AI stocks.

⍐