ADNOC Signs OpenAI Deal to Run Oil Field Planning on AI Agents, Hosted on G42 Servers in the UAE

Since ADNOC, G42 and OpenAI signed their agreement on Monday, Nov. 2, at the ADIPEC energy conference in Abu Dhabi, the contours of the deal have come into focus. It makes Abu Dhabi National Oil Company the first energy firm to deploy OpenAI's Frontier platform, a system for building and running fleets of AI agents that coordinate work across a large organization.

The models come from OpenAI. The computing sits on G42's sovereign AI infrastructure inside the UAE. ADNOC says its proprietary industrial data stays there too.

What ADNOC says it is buying

The first target is field master planning, the process of deciding where to drill, how to manage reservoirs and which surface facilities to build. ADNOC said OpenAI's reasoning models will pull together decades of data across thousands of wells, multiple stacked reservoirs and hundreds of surface facilities, along with capital, cost and commercial considerations.

Today, ADNOC said, that work runs as a sequence of separate subsurface, facilities and economic studies. The plan is to turn it into "a single, integrated decision." Crypto Briefing put the stated goal at compressing full-field planning from months to hours. That timeline is a goal, not a result.

Later phases are meant to reach production, processing, refining, logistics, sales and trading. ADNOC described interconnected agents that continuously forecast crude supply, demand, prices and margins.

The company says human experts will keep oversight of "consequential decisions."

Built on ENERGYai

This is not ADNOC's first AI project. Its ENERGYai platform, built with G42, AIQ and Microsoft, is already deployed across the upstream value chain and was trained on more than 50 years of ADNOC data, according to ADNOC.

ADNOC says ENERGYai builds deep knowledge of reservoirs and assets, while Frontier will connect that knowledge across the business so operational, capital and commercial decisions "can be made together rather than one after another."

Sultan Al Jaber, who is both ADNOC's CEO and the UAE's minister of industry and advanced technology, framed it for the ADIPEC audience: "From the moment ChatGPT was born, the world has asked what energy can do for AI. And now, together, we will find out what AI can do for energy."

Sam Altman, OpenAI's CEO, said the company's most advanced models would take on "the hardest, most valuable problems" ADNOC faces. He also pointed to the limit on the whole industry: "atoms are still atoms," and the physical world that powers AI will be the dominant factor in cost and abundance.

A wider Abu Dhabi push

The Abu Dhabi Media Office said OpenAI also signed agreements with Mubadala and with Inception42, a G42 unit, to apply AI to investment analysis, opportunity identification and risk assessment. OpenAI is one of the partners in the first phase of a 5-gigawatt AI data center project the UAE is building with American tech companies.

Reuters, citing sources, reported that OpenAI is racing to line up investor commitments, including from Abu Dhabi-based MGX, in a fundraising that could value the company above $1.4 trillion. Reuters tied the timing to Anthropic's expected public debut in mid-November. Anthropic has filed its S-1 with the SEC, with a listing possible as early as the week of Nov. 16.

The security backdrop

Altman flew in while the Gulf was on edge. ExxonMobil and Chevron cancelled plans to send executives to ADIPEC the week before, citing security concerns. BP, Eni and Baker Hughes executives attended anyway.

President Donald Trump had said he would not attack Iran before the Nov. 3 midterms, but on Friday, Oct. 30, he said Washington may carry out another military operation against Iran. Fox News Digital, at the opening ceremony, described the war involving Iran and disruptions to the Strait of Hormuz as placing "extraordinary pressure" on global energy supplies.

Dr. Angela Wilkinson, secretary general and CEO of the World Energy Council, told Fox News Digital the most encouraging part of the presentations was their focus on people: "It isn't about AI taking decisions for humanity, it's about AI enabling people to make better decisions."

The emissions argument

Not everyone buys the efficiency pitch. AI advocates argue that over the long run AI will save more energy than it consumes. A recent paper in NPJ Climate Action, cited by OilPrice, found that if fossil fuels gain economically anywhere above zero from AI integration, the emissions AI enables will exceed the emissions it avoids.

Semafor described the ADNOC agreement as the highest-profile example in a broader push at ADIPEC by tech companies to pitch AI tools for drilling rigs, refineries and other oil assets.

For U.S. policy, the fact pattern is plain. An American AI company is wiring its models into a Gulf state's oil operations, on Emirati servers, at a moment when Washington is weighing military action against a neighbor across the water. ADNOC and OpenAI have not published financial terms or a rollout timeline for the later production, refining and trading phases.