🔍 Read the full analysis: OpenAI Could Turn To Google TPUs, But Neither Company Has Confirmed It on ThorstenMeyerAI.com
Prime made for students and young adults
- Fast, free delivery for dorm and study essentials
- Prime Video and Amazon Music included
- Member-only deals
TL;DR
The Information and Bloomberg reported that OpenAI plans to buy Google’s AI chips, citing one person familiar with the matter each. Neither company has confirmed a purchase, and its size, price and timing remain unknown. OpenAI’s TPU engineering job and Google’s new sales push show why the report is plausible, but they do not establish that a deal has been signed.
The Information and Bloomberg reported this week that OpenAI plans to buy Google’s tensor processing units (TPUs), potentially adding another supplier to the infrastructure used to develop and run its AI systems. Neither company has confirmed a purchase: OpenAI declined to comment, Google did not comment, and the reported deal’s terms and timing are unknown.
The Information reported the plan on October 27, citing one person familiar with the matter. Reuters, relaying that report, said it could not learn financial terms or other details and reported that OpenAI declined to comment while Google did not immediately respond. Bloomberg reported the following day that OpenAI intends to buy Google AI computing chips as part of its AI build-out. It also cited one person familiar with the matter and said final terms were not known.
There is evidence that OpenAI is preparing to work with TPUs, but it does not confirm a purchase. An OpenAI careers listing published October 20 seeks a software engineer to work on TPU compilers and optimize the company’s frontier models on the chips. The role indicates technical work on TPU support; it is not a purchase order or proof of a large-scale deployment.
Neither outlet disclosed the number or generation of chips, the price, a delivery date, or where they would operate. The chips could potentially run in Google data centers, OpenAI facilities or through a cloud partner, but the supplied reporting does not establish which arrangement, if any, is planned.
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
A New Chip Supplier Could Shift Leverage
If a purchase goes ahead, Google would become another potential source of computing capacity for OpenAI, which has relied heavily on Nvidia while announcing other chip arrangements. A broader supplier base could give a large buyer more options when negotiating prices, delivery schedules and access to scarce capacity. Those are possible commercial effects, not confirmed outcomes of this reported plan.
The development also matters because Google is both a supplier of AI infrastructure and a competitor to OpenAI in consumer and enterprise AI. A deal would show that competition between the companies does not prevent them from doing business with each other. OpenAI said on October 21 that its frontier models would be available through Google Cloud, another sign of a growing commercial relationship, though that announcement does not establish a TPU purchase.
For ChatGPT users, there is no confirmed immediate change. The report does not say whether any chips would be used for model training, serving user requests or other workloads. Nor does it establish that a purchase would lower prices, make ChatGPT faster or change the service’s availability.
Popular right now
De'Longhi Magnifica XS ECAM22110B Automatic Espresso Machine
Philips 2200 Series Fully Automatic Espresso Machine with Classic Milk Frother
Jura E6 Automatic Espresso Machine, Piano Black (15622)
As an affiliate, we earn on qualifying purchases.
Google Expands TPU Sales Beyond Cloud
Google has begun offering TPUs outside its own data centers, a change that makes a reported OpenAI purchase more plausible than it would have been when the chips were mainly an internal resource. On its April 29, 2026, earnings call, Chief Executive Sundar Pichai said Google would deliver TPUs to selected customers’ own data centers in a hardware configuration. Google said on its October 28 earnings call that nearly 100 customers use TPUs and named Anthropic and Apple. It also announced partnerships with Oracle and Crux AI.
Oracle plans to offer Google TPUs through its cloud from 2027, according to the source material. Google’s commercial push is broader than a potential OpenAI arrangement: the chips are becoming an option for companies that want access to Google’s hardware beyond Google-operated facilities. The exact form and scale of any OpenAI arrangement remain unreported.
OpenAI has also announced agreements involving several non-Nvidia suppliers. The source material lists a 2025 letter of intent with Nvidia for 10 gigawatts of capacity, an AMD agreement for 6 gigawatts, 10 gigawatts of Broadcom custom chips, 750 megawatts from Cerebras and about 2 gigawatts of Amazon Trainium. These figures describe different announced arrangements and capacity measures; they do not show how many chips are currently operating or available to OpenAI. The Google report is therefore not OpenAI’s first reported move beyond Nvidia.
There is earlier evidence of limited TPU testing. Reuters reported in June 2025 that OpenAI had started renting Google TPUs through Google Cloud. OpenAI told Reuters at the time that it was in early testing with some TPUs and had no active plans to deploy them at scale. That history makes the current report a possible escalation, but does not confirm that a new purchase has been agreed.
“As TPU demand grows from AI labs, capital markets firms, and high performance computing applications, we will begin to deliver TPUs to a select group of customers in their own data centers in a hardware configuration.”
— Sundar Pichai, Google chief executive, on the company’s April 29, 2026, earnings call
Deal Size and Deployment Remain Unknown
No purchase has been confirmed by either company. The reports each rely on one person familiar with the matter, and the source material provides no public contract, company announcement or financial filing verifying a deal. The reporting should be treated as a report of intent, not proof that an agreement has been signed or completed.
Key details are missing: the chip quantity and generation, the cost, delivery schedule, intended workloads and hosting location. It is also unclear whether the reported plan would involve a direct hardware sale, cloud access or another arrangement. OpenAI’s hiring for TPU compiler work confirms technical interest, but does not resolve those questions.
The impact on OpenAI’s reliance on Nvidia is also uncertain. The company has announced several other chip agreements, but the available information does not show how much capacity has been delivered, when it will be usable or what share of OpenAI’s computing needs it could cover. Claims that this would end an Nvidia “lock” or immediately change ChatGPT performance go beyond what is confirmed.
Watch for Company Terms and Delivery Dates
The next meaningful confirmation would be a statement from OpenAI or Google, or a filing that identifies the arrangement’s scope. Further reporting could clarify whether the companies have reached final terms and whether the chips would be supplied directly or accessed through a cloud provider. Until then, price, scale and timing remain open questions.
Google’s Oracle arrangement is expected to begin in 2027, according to the source material, while other OpenAI chip commitments are also tied to future capacity. Readers should distinguish those announced plans from deployed systems: the reported Google purchase has not been confirmed, and there is no stated date for any OpenAI TPU delivery. For users, the available information points to no immediate change in ChatGPT.
Key Questions
Has OpenAI confirmed it will buy Google TPUs?
No. The Information and Bloomberg reported that OpenAI plans to buy the chips, but OpenAI declined to comment and Google did not confirm a purchase.
What evidence suggests OpenAI is working with TPUs?
OpenAI’s October 20 careers listing seeks an engineer to optimize frontier models on TPUs. Reuters also reported in 2025 that OpenAI was testing some TPUs through Google Cloud. Neither item proves a new purchase or a large-scale deployment.
Would this be OpenAI’s first move beyond Nvidia?
No. OpenAI has announced arrangements involving AMD, Broadcom, Cerebras and Amazon Trainium, in addition to its Nvidia relationship. The reported Google plan would add another potential supplier, but its status and scale are unknown.
Will Google TPUs make ChatGPT faster or cheaper?
There is no confirmed information showing that. The reports do not identify which workloads the chips would support or say whether any cost or performance change would reach ChatGPT users.
Source: ThorstenMeyerAI.com
