Technology · Global

OpenAI plans to buy Google AI chips, The Straits Times reports

The Straits Times reports that OpenAI intends to buy AI computing chips from Google, according to a person familiar with the matter. The deal’s terms are not known, and OpenAI has separately announced plans to use its own chips later in 2026.

AI-assisted synthesis · Based on 5 independent reports from 6 outlets · Article text read from 1 of 5 reportsHow it's made

View timeline · 7 entries

6 outlets2 regionsLatest reporting Oct 28, 2:28 AM UTC2 min read

Coverage: Center OnlyL 0 · C 5 · R 0 · 1 unrated

File photo

Photo: Coolcaesar / Wikimedia Commons (CC BY 4.0)

What happened

Key facts

A person familiar with the matter said OpenAI intends to buy AI computing chips from Google (per The Straits Times, investing.com, CNA).
The terms of a possible deal between OpenAI and Google are not yet known (per The Straits Times, investing.com, CNA).
OpenAI previously announced plans to begin using its own chips to support its AI models later in 2026 (per The Straits Times).
OpenAI is developing Jalapeno chips with Broadcom (per The Straits Times).
Oracle announced on Oct 27 that Google’s tensor processing units would be available through its cloud-computing infrastructure (per The Straits Times).
Google executive Amin Vahdat said AI’s next generation would require a diverse range of computing resources (per The Straits Times).

The brief

OpenAI intends to buy artificial-intelligence computing chips from Google, The Straits Times reported, citing a person familiar with the matter. The person said the companies had not settled the terms of a possible deal. The report describes a planned purchase, not a completed transaction.

The source places the possible purchase alongside OpenAI’s previously announced plan to begin using its own chips to support its AI models later in 2026. It identifies those chips as Jalapeno, which OpenAI is developing with Broadcom. The account does not say how many chips OpenAI might acquire from Google or give a price or purchase schedule.

Separately, Oracle announced on Oct 27 that Google’s tensor processing units, or TPUs, would be available through Oracle’s cloud-computing infrastructure. The announcement creates another described route for customers to access Google’s chips: through Oracle’s cloud.

The Straits Times does not say that OpenAI plans to obtain chips through Oracle, or that the cloud arrangement is part of OpenAI’s possible purchase. Google executive Amin Vahdat described the company’s view of computing needs in a statement announcing the Oracle arrangement. “The next generation of AI will require a diverse range of compute,” he said.

That is Google’s stated position on the needs of future AI systems; the account does not attribute a similar explanation for the possible purchase to OpenAI.

The report thus describes three distinct developments: OpenAI’s intended purchase of Google chips, its announced plan to use its own chips later in 2026, and Oracle’s announcement that it will offer Google TPUs through its cloud infrastructure. OpenAI’s chips are being developed with Broadcom, while the possible Google-chip purchase remains without disclosed terms.

The Straits Times attributes the purchase plan to a person familiar with the matter who requested anonymity because the discussions are private. The account does not identify that person or report a statement from OpenAI or Google confirming the proposed purchase. It also does not establish whether OpenAI’s own-chip plans and the possible Google purchase are linked.

Why it matters

OpenAI’s reported plan involves buying AI chips from Google, while the terms of a possible deal remain unknown.
OpenAI has separately announced that it plans to use its own chips later in 2026; it is developing Jalapeno with Broadcom.
Oracle’s announced cloud arrangement offers access to Google TPUs through Oracle’s infrastructure, a separate route described in the report.
Google executive Amin Vahdat said future AI would require a diverse range of computing resources.

What to watch next

Whether OpenAI and Google settle terms for the planned chip purchase.
Whether OpenAI begins using its own chips for AI models later in 2026, as previously announced.
Whether Oracle makes Google TPUs available through its cloud-computing infrastructure as announced.
Whether OpenAI or Google publicly confirms the proposed purchase.

How it's being covered

Coverage analysisShow details

6 source articles · 6 unique outlets · 5 distinct reports

Outlet political mix: Center Only· 5 Center · 1 unrated

Where the outlets are based: US/Canada 33% · Asia 33% · Other 34%

Outlet political mix

Center Only

Left

Center

Right

0 outlets

5 outlets

0 outlets

The marker shows coverage balance: each unique outlet counts once. 1 unrated outlet is not included.

Geography

Where the outlets covering this story are based.

6 unique outletsDominant coverage: US/Canada

All6 US/CA2 · 33%Asia2 · 33%Global2 · 33%

Coverage imbalance

Coverage is currently concentrated among Center-rated outlets (5 of 5 rated sources); comparatively few Left or Right-rated outlets have published on the event.

How we measure coverage

One source is from a wire service (AP, Reuters, AFP or Bloomberg), whose reporting other outlets often republish.

How outlets' accounts compare

Updat3 does not yet have enough full reports to compare how outlets framed this story.

Timeline

Developments from this story's sources, dated background we compiled and fact-checked, and our earlier coverage. Each row says which.

1998
Larry Page and Sergey Brin founded Google.
Background · compiled and fact-checked by Updat3
Dec 2015
OpenAI, founded in December 2015, built its AI services using large-scale computing infrastructure supplied by cloud and chip companies.
Background · from our historical context
Dec 2015
Sam Altman, Greg Brockman and other co-founders launched OpenAI as a research organization.
Background · compiled and fact-checked by Updat3
May 2016
Google announced its first TPU in May 2016, describing it as a custom processor for machine-learning workloads.
Background · from our historical context
Jul 2019
In July 2019, Microsoft announced a $1 billion investment in OpenAI and a partnership under which Microsoft Azure would provide computing resources.
Background · from our historical context
Jan 2023
In January 2023, the companies expanded that partnership, with Microsoft committing to continue providing Azure infrastructure.
Background · from our historical context
Oct 27, 2026
On October 27, 2026, Oracle announced that Google’s tensor processing units (TPUs) would be available through Oracle Cloud Infrastructure, adding a new route for customers to access Google-designed AI chips.
Background · from our historical context

How we got here

On October 27, 2026, Oracle announced that Google’s tensor processing units (TPUs) would be available through Oracle Cloud Infrastructure, adding a new route for customers to access Google-designed AI chips. Google announced its first TPU in May 2016, describing it as a custom processor for machine-learning workloads.

The move reflected a broader shift toward specialized chips: Nvidia’s graphics processors had become central to training large AI models, while cloud providers began developing their own alternatives. OpenAI, founded in December 2015, built its AI services using large-scale computing infrastructure supplied by cloud and chip companies.