South Korean start-up Rebellions Inc yesterday launched an artificial intelligence (AI) chip, racing to win government contracts as Seoul seeks a place for local companies in the exploding industry.
The company’s Atom chip is the latest South Korean attempt to challenge global leader Nvidia Corp in the hardware that powers the potentially revolutionary AI technology.
AI is the talk of the tech world, as ChatGPT — a chatbot from Microsoft Corp-backed OpenAI that generates articles, essays, jokes and even poetry — has become the fastest-growing consumer app in history just two months after launch.
Photo: Courtesy of Rebellions Inc via Reuters
Nvidia, a US chip designer, has a commanding share of high-end AI chips, making up about 86 percent of the computing power of the world’s six biggest cloud services as of December last year, Jefferies Group LLC chip analyst Mark Lipacis said.
The South Korean government wants to foster a domestic industry, investing more than US$800 million over the next five years for research and development in a bid to lift the market share of South Korean AI chips in domestic data centers from essentially zero to 80 percent by 2030.
“It’s hard to catch up to Nvidia, which is so far ahead in general-purpose AI chips,” Korea Institute for Industrial Economics and Trade senior researcher Kim Yang-paeng said. “But it’s not set in stone, because AI chips can carry out different functions and there aren’t set boundaries or metrics.”
Rebellions’ Atom is designed to excel at running computer vision and chatbot AI applications.
Because it targets specific tasks rather than doing a wide range, the chip consumes only about 20 percent of the power of an Nvidia A100 chip on those tasks, Rebellions cofounder and chief executive Park Sung-hyun said.
A100 is the most popular chip for AI workloads, powerful enough to create — in industry lingo, “train” — the AI models. Atom, designed by Rebellions and manufactured by South Korean giant Samsung Electronics Co, does not do training.
While countries such as Taiwan, China, France, Germany and the US have extensive plans to support their semiconductor companies, the South Korean government is rare in singling out AI chips for a concentrated push.
Seoul is to put out a notice this month for two data centers, called neural processing unit farms, with only domestic chipmakers allowed to bid, an official at the South Korean Ministry of Science and Information Communications Technology said.
In a country whose firms supply half the world’s memory chips, the authorities want to create a market that can be a test bed for AI chipmakers, aiming to foster global competitors.
“The government is twisting the arm of the data centers and telling them: ‘Hey, use these chips,’” said Park, a former Morgan Stanley engineer.
Without such support, data centers and their customers would likely stick with Nvidia chips, he said.
Sapeon Korea Inc also plans to participate in the project, the SK Telecom Co subsidiary said.
FuriosaAI, backed by South Korea’s top search engine Naver Corp and state-run Korea Development Bank, said it would also bid.
“There’s a lot of momentum behind Nvidia’s developments. These start-ups have got to build momentum, so that will take time,” Gartner Inc analyst Alan Priestley said. “But government incentives such as what’s happening in Korea could well affect the market share within Korea.”
Rebellions would seek to participate in the government project in a consortium with KT Corp — a big South Korean telecom, cloud and data center operator — in the hopes of weaning Nvidia customers off the US supplier.
“Amid high dependence on foreign GPUs [graphics processing units] globally, the cooperation between KT and Rebellions will allow us to have an ‘AI full stack’ that encompasses software and hardware based on domestic technology,” KT vice president Bae Han-chul said.
Rebellions declined to give a forecast for its AI chip venture. It has raised 122 billion won (US$95.47 million), including 30 billion won from KT in a funding round joined by Singapore’s Temasek Pavilion Capital and a 10 billion won grant from the South Korean government.
Shiina Ito has had fewer Chinese customers at her Tokyo jewelry shop since Beijing issued a travel warning in the wake of a diplomatic spat, but she said she was not concerned. A souring of Tokyo-Beijing relations this month, following remarks by Japanese Prime Minister Sanae Takaichi about Taiwan, has fueled concerns about the impact on the ritzy boutiques, noodle joints and hotels where holidaymakers spend their cash. However, businesses in Tokyo largely shrugged off any anxiety. “Since there are fewer Chinese customers, it’s become a bit easier for Japanese shoppers to visit, so our sales haven’t really dropped,” Ito
The number of Taiwanese working in the US rose to a record high of 137,000 last year, driven largely by Taiwan Semiconductor Manufacturing Co’s (TSMC, 台積電) rapid overseas expansion, according to government data released yesterday. A total of 666,000 Taiwanese nationals were employed abroad last year, an increase of 45,000 from 2023 and the highest level since the COVID-19 pandemic, data from the Directorate-General of Budget, Accounting and Statistics (DGBAS) showed. Overseas employment had steadily increased between 2009 and 2019, peaking at 739,000, before plunging to 319,000 in 2021 amid US-China trade tensions, global supply chain shifts, reshoring by Taiwanese companies and
Taiwan Semiconductor Manufacturing Co (TSMC) Chairman C.C. Wei (魏哲家) and the company’s former chairman, Mark Liu (劉德音), both received the Robert N. Noyce Award -- the semiconductor industry’s highest honor -- in San Jose, California, on Thursday (local time). Speaking at the award event, Liu, who retired last year, expressed gratitude to his wife, his dissertation advisor at the University of California, Berkeley, his supervisors at AT&T Bell Laboratories -- where he worked on optical fiber communication systems before joining TSMC, TSMC partners, and industry colleagues. Liu said that working alongside TSMC
TECHNOLOGY DAY: The Taiwanese firm is also setting up a joint venture with Alphabet Inc on robots and plans to establish a firm in Japan to produce Model A EVs Manufacturing giant Hon Hai Precision Industry Co (鴻海精密) yesterday announced a collaboration with ChatGPT developer OpenAI to build next-generation artificial intelligence (AI) infrastructure and strengthen its local supply chain in the US to accelerate the deployment of advanced AI systems. Building such an infrastructure in the US is crucial for strengthening local supply chains and supporting the US in maintaining its leading position in the AI domain, Hon Hai said in a statement. Through the collaboration, OpenAI would share its insights into emerging hardware needs in the AI industry with Hon Hai to support the company’s design and development work, as well
RSS