After Nvidia, Mediatek may have convinced another huge tech company to use its expertise to develop AI chips

News
By published

MediaTek’s AI push strengthens with Google TPU collaboration deal

MediaTek
Are the price of flagships about to drop? (Image credit: Future)
  • MediaTek challenges Broadcom in Google's evolving AI hardware strategy
  • Next-gen TPU aims for better AI processing and lower costs
  • Google diversifies AI chip supply, reducing reliance on Broadcom

MediaTek and Google will collaborate to develop the next-gen tensor processing units (TPUs), a move which means Google will likely be reducing its reliance on Broadcom in its AI tools hardware business strategy.

The aim of this collaboration is to design an affordable and energy-efficient TPU, produced by TSMC and will likely launch as part of Google’s 7th-gen TPU series.

While Google will lead the overall design, MediaTek will handle input/output modules to facilitate communication between the processor and peripherals, marking a shift from Broadcom’s previous role in TPU core development, though Google is not cutting ties with Broadcom entirely.

Developing more efficient AI processors

Google’s TPUs, designed for demanding AI computations and essential for training the best LLMs, will reportedly see improved efficiency and performance, addressing the increasing demand for powerful AI processing.

Google has been developing its own AI server chips for internal use and cloud services to reduce reliance on external providers like Nvidia, whose dominance in AI hardware contrasts with Google’s push for a more self-sufficient ecosystem through TPU investments.

Meanwhile, Mediatek has been expanding its presence in AI chip development, including partnerships in AI supercomputers.

Ine arly 2025, the company announced an AI supercomputer project collaboration, with its latest partnership with Google aligning with its broader strategy to strengthen its position in the AI industry.

If this collaboration is successful, it may influence the development of the best LLM models by making AI processing more efficient.

Via UDN

You may also like

Efosa Udinmwen
Efosa Udinmwen
Freelance Journalist

Efosa has been writing about technology for over 7 years, initially driven by curiosity but now fueled by a strong passion for the field. He holds both a Master's and a PhD in sciences, which provided him with a solid foundation in analytical thinking. Efosa developed a keen interest in technology policy, specifically exploring the intersection of privacy, security, and politics. His research delves into how technological advancements influence regulatory frameworks and societal norms, particularly concerning data protection and cybersecurity. Upon joining TechRadar Pro, in addition to privacy and technology policy, he is also focused on B2B security products. Efosa can be contacted at this email: udinmwenefosa@gmail.com

You must confirm your public display name before commenting

Please logout and then login again, you will then be prompted to enter your display name.