🔍 Read the full analysis: OpenAI And Google’s TPUs: A Potential Alternative To Nvidia Chips? on ThorstenMeyerAI.com
Get everyday essentials delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
The Information and Bloomberg reported that OpenAI plans to buy Google’s AI chips, known as TPUs, but the companies have not confirmed a deal and its terms are unknown. Google’s expanding TPU sales and OpenAI’s existing agreements with several chipmakers point to a broader effort to diversify computing supply, though no near-term change for ChatGPT users has been announced.
OpenAI reportedly plans to buy Google’s tensor processing units (TPUs), according to separate reports by The Information on October 27 and Bloomberg on October 28. Neither company has confirmed a purchase, and the price, chip quantity and delivery schedule are unknown, but the report comes as Google begins selling TPUs beyond its own data centers and OpenAI pursues computing capacity from several suppliers.
The Information and Bloomberg each cited one person familiar with the matter in reporting OpenAI’s plans. Reuters, which relayed The Information’s report, said it could not learn financial terms or other details. OpenAI declined to comment; Google did not comment, according to the source material. The reports indicate a plan, not a completed purchase or a confirmed supply agreement.
There is separate evidence that OpenAI is working on TPU software. A TPU compiler engineering job posted on OpenAI’s careers page on October 20 seeks someone to optimize the company’s frontier models on TPUs. The posting supports the conclusion that OpenAI is doing technical work with the hardware, but it does not establish that the company has placed an order.
Google, meanwhile, has started offering TPUs outside its own infrastructure. On its October 28 earnings call, the company said nearly 100 customers use TPUs and named Anthropic and Apple. Oracle announced it plans to offer Google TPUs through its cloud from 2027. Google’s broader shift gives prospective buyers another way to access the chips, though the OpenAI reports do not say where its TPUs would run.
Infographic 01 · Reported, documented, confirmed
OpenAI and Google’s TPUs: what we actually know
The purchase itself is a report. The pieces around it, Google selling its chips and OpenAI hiring for them, are on the record.
- OpenAI plans to buy Google’s AI chipsReported
The Information, 27 October; Bloomberg, 28 October. Each cites one person familiar with the matter. OpenAI declined to comment; Google did not comment.
- OpenAI is hiring to optimize its frontier models on TPUsPrimary document
“Software Engineer, TPU Compilers”, posted on OpenAI’s careers page on 20 October 2026. A sign of serious work, not a purchase order.
- Google sells TPU systems into customers’ own data centersConfirmed
Announced on 29 April, first deliveries in the second quarter, “a meaningful revenue ramp in 2027” (Alphabet’s CFO, 28 October).
- Oracle will offer Google TPUs in its cloudConfirmed
The first cloud outside Google to do so. Available in calendar year 2027, ramping over several years.
- Price, number of chips, TPU generation, delivery dates, locationNot known
Nothing published. Not even whether the chips would sit in Google’s data centers, OpenAI’s or a partner’s.
How OpenAI and Google got here
- Jun 2025OpenAI rents Google TPUs, calls it early testingReuters: no active plans to deploy them at scale
- 29 Apr 2026Google: TPUs for customers’ own data centersAlphabet Q1 earnings call
- 20 Oct 2026OpenAI posts a TPU compiler jobOpenAI careers page
- 21 Oct 2026OpenAI models coming to Google CloudOpenAI announcement, as reported by Reuters
- 27 Oct 2026The Information reports the purchase plan; Oracle announces TPUsReported · Oracle press release
- 28 Oct 2026Bloomberg reports it too; Alphabet’s call does not mention OpenAIReported · Alphabet Q3 call
“Reported” means the outlets’ own reporting, which I could read only in part (paywalls). “Confirmed” means the company’s own statement or filing.
Sources: The Information via Reuters, 27 Oct 2026, Bloomberg, 28 Oct 2026, as cited by The Asia Business Daily, OpenAI careers, posted 20 Oct 2026, Reuters, 30 Jun 2025, Alphabet Q1 call, 29 Apr 2026, Q3 call, 28 Oct 2026, Oracle, 27 Oct 2026.
Powered by Thorsten Meyer AI
More Options for AI Computing
If the reported purchase proceeds, it would give OpenAI another potential source of computing capacity as demand for training and running AI models grows. The development could matter beyond the two companies: more competition among chip suppliers may strengthen large buyers’ ability to negotiate prices, access and delivery schedules. That is a possible bargaining effect, not a confirmed reduction in costs.
The report also brings together two companies that compete in AI products while building commercial ties. OpenAI said on October 21 it would make its frontier models available through Google Cloud. A TPU purchase would add a potential hardware relationship to that arrangement, but the details and scale remain unverified.
For users of ChatGPT, there is no announced change to the service. The source material says many of OpenAI’s announced chip arrangements involve capacity expected in 2027 or later. It is not known whether a Google deal would affect ChatGPT’s speed, availability or price, or when any impact might occur.
Products Worth Considering
DJI Osmo Mobile 7P Phone Gimbal with Mic Mini Transmitter, Infinity Black
Jackery Explorer 2000 v2 Portable Power Station with Two 200W Solar Panels
Bose QuietComfort Ultra Wireless Noise-Cancelling Headphones with Spatial Audio, Black
As an affiliate, we earn on qualifying purchases.
OpenAI’s Expanding Chip Supplier List
The reported Google plan would not mark OpenAI’s first move beyond Nvidia. The company has announced arrangements involving AMD, Broadcom, Cerebras and Amazon, alongside Nvidia. The source material lists a 2025 letter of intent with Nvidia for 10 gigawatts, an AMD agreement for 6 gigawatts, 10 gigawatts of OpenAI-designed chips with Broadcom, a 750-megawatt Cerebras arrangement and about 2 gigawatts of Amazon Trainium capacity. These are reported as different kinds of agreements and plans; they do not establish that all the capacity is already delivered or operational.
Google’s chips were historically used primarily within its own data centers, although customers could rent them through Google Cloud. In April 2026, Chief Executive Sundar Pichai said Google would begin delivering TPUs to selected customers’ own data centers. The company has since named additional customers and partnerships, while Oracle has said it will offer the chips in its cloud starting in 2027.
OpenAI had also tested Google TPUs earlier. Reuters reported in June 2025 that the company had begun renting them through Google Cloud. OpenAI then told Reuters it was in early testing and had no active plans to deploy TPUs at scale. That earlier account shows that technical testing predates the current reports, but does not confirm the newly reported purchase plan.
Purchase Terms Remain Unknown
Neither OpenAI nor Google has confirmed the newly reported plan. The available reporting does not establish how many chips OpenAI might buy, which TPU generation is involved, what they would cost or when they could be delivered. It is also unknown whether the chips would operate in Google’s data centers, in OpenAI’s own facilities or through a cloud partner such as Oracle.
The two reports cite one person familiar with the matter each, and no public contract or purchase order has been identified in the source material. OpenAI’s job listing is evidence of TPU-related engineering work, not proof of a deal. It remains unclear whether the reported plan will become a binding agreement or how much computing capacity, if any, it would add.
Watch for a Confirmed Agreement
The next meaningful developments would be confirmation from OpenAI or Google, or further reporting that establishes the agreement’s scope, price and timing. Any announced deployment would also clarify where the chips will run and whether they are intended for model training, inference or both.
Google’s planned Oracle cloud availability from 2027 provides a separate milestone for TPU access outside Google’s own facilities. Until a purchase is confirmed and delivery details emerge, the reports describe a possible addition to OpenAI’s supply options—not a change to ChatGPT users’ current experience.
Key Questions
Has OpenAI confirmed that it will buy Google TPUs?
No. The Information and Bloomberg reported a plan, but neither OpenAI nor Google has confirmed a purchase in the source material. OpenAI declined to comment, and Google did not comment.
What are Google TPUs?
TPUs, or tensor processing units, are AI chips designed by Google. The company uses them in its own data centers and has begun offering them to selected customers outside those facilities.
Would a Google deal be OpenAI’s first move beyond Nvidia?
No. OpenAI has already announced arrangements involving AMD, Broadcom, Cerebras and Amazon, in addition to Nvidia. The reported Google plan would add another possible supplier, not mark OpenAI’s first non-Nvidia arrangement.
Will this make ChatGPT cheaper or faster?
There is no confirmed effect on ChatGPT’s price or performance. The reports do not provide delivery dates or explain how any Google chips would be used, so user impact cannot yet be assessed.
Source: ThorstenMeyerAI.com
