MediaTek AI ASIC Targets $16B Revenue in 2027

According to MediaTek CEO Rick Tsai, the company's artificial intelligence application-specific integrated circuit (AI ASIC) business is entering a period of rapid revenue growth, following the initial shipment of its first data center AI accelerator in September 2026. Speaking at the company's third-quarter earnings conference on October 30, Tsai said shipments are expected to ramp up significantly in the fourth quarter, with MediaTek confident of achieving its US$2.5 billion revenue target for the first AI ASIC project.

Looking ahead to 2027, MediaTek expects revenue from its first AI ASIC project to reach the upper end of its original target range, approximately US$16 billion. This outlook highlights the company's accelerating transformation from a leading smartphone chip developer into a major player in AI computing, spanning both cloud and edge applications.

Tsai noted that global demand for AI accelerators remains strong. MediaTek is working closely with supply chain partners to secure additional critical production capacity and meet growing customer requirements. The successful initial shipments and production ramp-up of its first AI ASIC represent an important milestone, demonstrating the company's capabilities in integrating advanced intellectual property (IP), leading-edge process technologies, advanced packaging, and global manufacturing resources to develop competitive customized chips for large-scale data centers.

Development of MediaTek's second AI accelerator project is also progressing as planned, with mass production expected to begin in the first quarter of 2028. Designed for customers' next-generation AI infrastructure, the new accelerator will incorporate more advanced design technologies to deliver several times higher computing performance per watt while significantly reducing data center total cost of ownership (TCO). The project is expected to become another important growth driver from 2028 onward.

To strengthen its position in AI infrastructure, MediaTek continues to invest in key technologies, including 448G SerDes, 3.5D advanced packaging, and co-packaged optics (CPO). The company is also expanding beyond individual semiconductor components into system-level and rack-scale solutions, helping customers simplify integration across different AI computing architectures.

MediaTek is further deepening its collaboration with NVIDIA through the NVLink Fusion platform, which enables customized CPUs to connect with NVIDIA's NVLink ecosystem. The company is also supporting the UALink architecture, giving customers more flexibility in designing AI computing racks while accelerating product deployment and revenue contributions. The partnership has expanded beyond cloud AI into personal computing and automotive applications.

In the edge AI market, a computing product jointly developed by MediaTek and NVIDIA, integrating NVIDIA GPUs with MediaTek SoCs, was launched in mid-October. Development of its second-generation SoC has also begun, further expanding the companies' collaboration in AI computing.

Meanwhile, MediaTek's flagship Dimensity 9500 processor has been selected for Samsung's Galaxy Tab S11 series AI tablets, marking the third consecutive generation of Samsung flagship tablets to adopt MediaTek's chips.

On the financing front, MediaTek disclosed that it completed an overseas convertible bond issuance during the third quarter, attracting participation from NVIDIA and Alphabet, Google's parent company. The fundraising is intended to strengthen strategic relationships within the AI ecosystem and improve access to critical component production capacity. NVIDIA had previously announced a US$3.5 billion investment in MediaTek through convertible bonds.

Tsai emphasized that MediaTek will continue expanding cooperation with major companies across the global AI industry, extending its capabilities from customized semiconductor development to comprehensive AI infrastructure solutions. The company also plans to pursue new customer opportunities and strengthen its presence in data centers, physical AI, and automotive computing, positioning these markets as long-term growth drivers.