After the Chinese business DeepSeek’s artificial intelligence (AI) model caused a stir on Wall Street, US President Donald Trump has referred to the company’s rise as “a wake-up call” for the US tech sector.
Major tech companies like Nvidia saw a dramatic decline in share prices, and the chip giant lost about $600 billion (£482 billion) in market value.
DeepSeek’s assertion that its R1 model was created at a fraction of the price of its competitors has rocked the industry and raised concerns about the extent of expenditures that US companies are planned as well as the future of America’s dominance in AI.
Within a week of its release, DeepSeek has risen to the top of the US app download charts.
Trump responded to the news by suggesting that recent advancements in China’s AI sector could be “a positive” for the US.
“If you could complete the task more affordably, for less money, and still achieve the same outcome.” On board Air Force One, he told reporters, “I think that’s a good thing for us.”
The US will continue to dominate the field, he added, adding that he was not worried about the breakthrough.
However, DeepSeek has raised cyber security concerns in some countries with Australian science minister Ed Husic urging caution.
He told Australia’s national broadcaster ABC: “There are a lot of questions that will need to be answered in time on quality, consumer preferences, data and privacy management.”
DeepSeek is powered by the open source DeepSeek-V3 model, which its researchers claim was trained for about $6 million (£4.2 million), which is significantly less than the billions spent by competitors. However, other AI experts have disputed this claim. DeepSeek’s emergence coincides with the US limiting the sale of the advanced chip technology that powers AI to China. Chinese AI developers have shared their work with one another and experimented with new approaches to the technology in order to continue working without steady supplies of imported advanced chips. This has led to AI models that require far less computing power than previously thought possible, which also means they cost a lot less than previously thought possible, which has the potential to upend the industry.
The FTSE 100 stock index of the largest publicly traded businesses in the UK looked sturdy in early trading on Tuesday, increasing by 0.46% after Monday’s shock to US markets.
The tech-heavy Nasdaq index’s futures were also up 0.1% following a minor after-hours increase in Nvidia’s stock.
However, shares of Japanese AI companies such as Tokyo Electron, Softbank, and Advantest plummeted, contributing to a 1.4% decline in the benchmark Nikkei 225.
For the Lunar New Year holiday, a number of other Asian markets are closed. The Mainland The financial markets in China will close on Tuesday and reopen on February 5.
Who founded DeepSeek?
Liang Wenfeng established the business in Hangzhou, a city in southeast China, in 2023.
A graduate of information and electronic engineering, the 40-year-old also created the hedge fund that supported DeepSeek.
He was recently spotted at a meeting between Chinese Premier Li Qiang and industry specialists.
Speaking to The China Academy in July 2024, Mr. Liang expressed amazement at the response to his AI model’s initial iteration.
“We didn’t expect pricing to be such a sensitive issue,” he stated.
“We were simply following our own pace, calculating costs, and setting prices accordingly.”
Following the release of DeepSeek-R1 earlier this month, the business advertised “performance on par with” one of OpenAI’s most recent models for tasks like coding, math, and natural language thinking.
Prominent individuals, such as OpenAI chief Sam Altman, have commended DeepSeek’s technology, calling it “an impressive model, particularly around what they’re able to deliver for the price.” Altman also stated that OpenAI would “obviously deliver much better models” going forward.
According to Marina Zhang, an associate professor at the University of Technology Sydney who studies China’s high-tech industries, “DeepSeek’s ability to rival US models despite limited access to advanced hardware demonstrates that software ingenuity and data efficiency can compensate for hardware constraints.”
Ion Stoica, co-founder and executive chair of Databricks, a company that develops AI software, told the BBC that DeepSeek’s reduced price may encourage more businesses to use AI.
ICYMT: New NPA Boss vows sector reforms and innovation
“If that happens, this reduction in cost can accelerate the progress of AI,” he stated. “So overall, the market will expand faster, and the value of the market will grow faster.”
According to the Chinese company, its model can be trained on 2,000 specialized chips, whereas the industry standard is 16,000 chips.
However, not everyone is persuaded. Some people, including tech tycoon Elon Musk, have questioned some of DeepSeek’s assertions.
