Google plans to more than double production of its TPUs by 2028, analysts predict, as it steps up its ambitions for a processor that chip consultancy SemiAnalysis says is now \u201cneck and neck with king of the jungle Nvidia\u201d for building and running cutting-edge AI systems.\u00a0<\/p>\n
Nvidia investors have also been rattled by the prospect of Google offering TPUs to customers beyond its own cloud computing platform. This includes a recent deal to provide AI start-up Anthropic with 1mn TPUs, worth tens of billions of dollars. <\/p>\n
Google argues vertical integration \u2014 developing AI hardware, software and chips largely in-house \u2014 will deliver both technical advantages and huge profits. <\/p>\n
Gemini 3, like Google\u2019s previous models, was trained largely on TPUs. OpenAI relies mainly on Nvidia graphics processing units to build the large language models that power ChatGPT.\u00a0<\/p>\n
![\"Line](\"https:\/\/www.newsbeep.com\/ca\/wp-content\/uploads\/2025\/12\/https:\/\/d6c748xw2pzm8.cloudfront.net\/prod\/a8e269f0-d453-11f0-a248-c5f4f111f20c-standard.png\")
<\/p>\n
\u201cThe most important thing is . . . that full stack approach,\u201d said Koray Kavukcuoglu, Google\u2019s AI architect and DeepMind\u2019s chief technology officer, \u201cI think we have a unique approach there.\u201d <\/p>\n
Combining that with understanding how billions of consumers use Gemini, AI overviews in search and other products gives Google a huge advantage, he added.\u00a0<\/p>\n
Morgan Stanley estimates that every 500,000 TPUs sold to external customers could generate as much as $13bn in revenue for Google. The Big Tech company works primarily with chip design partner Broadcom, as well as with MediaTek, to develop the processors.\u00a0<\/p>\n
Morgan Stanley analysts also predict that Taiwan Semiconductor Manufacturing Company will produce 3.2mn TPUs next year, growing to 5mn in 2027 and 7mn in 2028. <\/p>\n
\u201cGrowth in 2027 is significantly stronger than previously anticipated,\u201d the bank\u2019s analysts said in a recent research note. <\/p>\n
Nvidia\u2019s stock fell sharply<\/a> last month following a report in The Information that Meta was in talks with Google to buy TPUs. Meta has declined to comment on the report.\u00a0<\/p>\n Some analysts believe Google could also strike such deals with OpenAI, Elon Musk\u2019s xAI or start-ups such as Safe Superintelligence, potentially driving upwards of $100bn in new Google revenues over the coming years.\u00a0<\/p>\n Experts add that AI-enabled coding tools could make it easier for potential TPU customers to rewrite their software, which until now has largely been built on top of Nvidia\u2019s proprietary Cuda platform. <\/p>\n Nvidia has sought to assuage market concerns, saying it was still \u201ca generation ahead of the industry\u201d and was \u201cthe only platform that runs every AI model\u201d, adding: \u201cWe continue to supply to Google.\u201d<\/p>\n The chipmaker added it offers \u201cgreater performance, versatility, and fungibility\u201d than processors such as TPUs, \u201cwhich are designed for specific AI frameworks or functions\u201d.<\/p>\n Recommended<\/p>\n
![\"Two](\"https:\/\/www.newsbeep.com\/ca\/wp-content\/uploads\/2025\/12\/https:\/\/images.ft.com\/v3\/image\/raw\/https%3A%2F%2Fd1e00ek4ebabms.cloudfront.net%2Fproduction%2F8a38dc.jpeg\")
<\/a><\/p>\n