MediaTek Inc (聯發科), the world’s largest mobile phone chip supplier, yesterday projected that revenue this year would grow by a mid-to-high-single-digit percentage, as robust demand for artificial intelligence (AI) accelerators would compensate for weakness in smartphone chips.
The Hsinchu-based chip designer said the mobile chip segment is facing strong headwinds this year, with global smartphone shipments forecast to fall about 15 percent annually due to a memorychip supply crunch and rising prices.
On the data center front, MediaTek said that with demand surging, it has significantly raised its revenue forecast for the global AI accelerator market to between US$70 billion and US$80 billion next year.
Photo: CNA
That is a dramatic step-up from its previous estimate of between US$60 billion and US$70 billion by 2028.
MediaTek has set a target of capturing 10 to 15 percent of that addressable market.
“Everybody can see that the demand for data center infrastructure, data center computing power just continues to grow and, if anything, continues to accelerate,” MediaTeck CEO Rick Tsai (蔡力行) told a virtual investor conference.
The company projects that its first AI application-specific IC (ASIC) project for a US hyperscaler would contribute US$2 billion in revenue in the fourth quarter of this year, doubling from its estimate of US$1 billion three months ago, and increase further next year.
Mediatek did not disclose the customer’s name, but reports said the company is designing a tensor process unit — an AI ASIC — for Google.
The design for another AI accelerator ASIC project is also under way, Tsai said.
Mass production of the second AI chip, which is more powerful and bigger than the first one, is targeted to start by the end of next year, Tsai said.
MediaTek is also engaged in several new data center ASIC projects, some of which are in the final stages of discussion, he said.
For this quarter, MediaTek expects revenue to be flat or slide 6 percent to NT$140.2 billion or NT$149.2 billion (US$4.43 billion and US$4.71 billion) from NT$149.15 billion last quarter, dragged by sagging smartphone chip demand. On an annual basis, revenue would drop by 1 to 17 percent, it said.
MediaTek expects its smartphone chip business to improve in the second half of this year, as it has secured several design-ins for its flagship chips.
The company’s first 2-nanometer smartphone chip produced by Taiwan Semiconductor Manufacturing Co (台積電) is to hit the market in the third quarter.
Smartphone chips accounted for 49 percent of the company’s total revenue last quarter. Chips used in smart edge devices, such as WiFi chips, networking and automotive chips, made up 46 percent.
Gross margin this quarter is expected to be between 44.5 percent and 47.5 percent, versus 46.3 percent last quarter, MediaTek said.
For the full year, gross margin is to hold steady at that range, it said.
MediaTek reported a net profit of NT$24.15 billion last quarter, down 17.6 percent from NT$29.33 billion a year earlier, but up 5.4 percent from NT$22.93 billion a quarter earlier.
Earnings per share dropped to NT$15.17 last quarter from NT$18.43 in the previous year, but increased sequentially from NT$14.39.
SECOND-RATE: Models distilled from US products do not perform the same as the original and undo measures that ensure the systems are neutral, the US’ cable said The US Department of State has ordered a global push to bring attention to what it said are widespread efforts by Chinese companies, including artificial intelligence (AI) start-up DeepSeek (深度求索), to steal intellectual property from US AI labs, according to a diplomatic cable. The cable, dated Friday and sent to diplomatic and consular posts around the world, instructs diplomatic staff to speak to their foreign counterparts about “concerns over adversaries’ extraction and distillation of US AI models.” Distillation is the process of training smaller AI models using output from larger, more expensive ones to lower the costs of training a powerful new
Shares of Taiwan Semiconductor Manufacturing Co (TSMC, 台積電) have repeatedly hit new highs, but an equity analyst said the stock’s valuation remains within a reasonable range and any pullback would likely be technical. The contract chipmaker’s historical price-to-earnings (P/E) ratio has ranged between 20 and 30, Cathay Futures Consultant Co (國泰證期) analyst Tsai Ming-han (蔡明翰) told Central News Agency. With market consensus projecting that TSMC would post earnings per share of about NT$100 (US$3.17) this year, supported by strong global demand for artificial intelligence (AI) applications, and the stock currently trading at a P/E ratio of below 25, Tsai said the valuation
The artificial intelligence (AI) boom has triggered a seismic reshuffling of global equity markets, with Taiwan and South Korea muscling past European nations one by one. With its stock market now valued at nearly US$4.3 trillion, Taiwan surpassed the UK, Europe’s biggest market, earlier this month, data compiled by Bloomberg showed. South Korea is about US$140 billion away from doing the same. The tech-heavy Asian markets have shot past Germany and France in the past seven months. The shift is largely down to massive gains in shares of three companies that provide essential hardware for AI: Taiwan Semiconductor Manufacturing Co (TSMC, 台積電),
NEXT-GENERATION: The company announced plans to roll out platforms with 100 TOPS performance and expects AI-capable systems to become the standard Industrial computer maker Advantech Co (研華) is eyeing strong growth potential in edge artificial intelligence (AI), as the technology shifts from cloud-centric computing to edge deployment and real-world applications, Advantech embedded Internet of Things president Miller Chang (張家豪) said on Thursday last week. Amid rising demand for inference as the industry shifts from training, edge AI is emerging as a key field, given the need for low latency, real-time response and data privacy, Chang said in an exclusive interview with the Taipei Times. While consumer devices such as smartphones already feature some AI capabilities, large-scale inference demand would mainly come from applications
RSS