ExxonMobil and Chevron posted sharply higher second-quarter profits on Friday after the U.S.-Iran conflict pushed crude oil prices higher, boosting earnings across their upstream and refining operations.

The results add to the spike in crude prices and tighter fuel markets that have strengthened the financial performance of the world’s largest oil producers, even as concerns persist over the security of global energy supplies.

Chevron delivered the stronger earnings surprise, while Exxon reported record production but narrowly missed Wall Street’s profit expectations.

Chevron Posts Nearly 400% Jump In Profit

Chevron reported net income of $12 billion for the second quarter, nearly five times the $2.5 billion earned in the same period last year.

Adjusted earnings came in at $6.06 per share, comfortably ahead of analysts’ expectations of $5.56 per share, according to LSEG.

Revenue rose to $70 billion, surpassing forecasts of $62 billion.

Chief Executive Mike Wirth said the company is benefiting from strong performance across its operations.

“We’re kind of firing on all cylinders, which is good, because the world needs it,” Wirth told CNBC.

He also warned that risks to global oil supplies extend beyond the Strait of Hormuz, noting that attacks by Iran-backed Houthi forces in the Red Sea are threatening another critical shipping corridor used for Saudi crude exports.

“The situation is under stress and I’m afraid it’s going to continue to do so,” Wirth said. “We’re running out of time. Every day that goes by, the situation gets more difficult.”

Exxon Doubles Quarterly Earnings

ExxonMobil reported second-quarter profit of $14.5 billion, roughly double the $7.1 billion earned a year earlier.

Adjusted earnings were $3.52 per share, slightly below analysts’ expectations of $3.60 per share, although revenue significantly exceeded forecasts.

Revenue climbed to $116 billion, well above the consensus estimate of $97.8 billion, reflecting higher oil prices and stronger production volumes.

Despite the earnings miss, Exxon continued to demonstrate operational strength through record output and improved refining performance.

The earnings surge followed a sharp increase in crude prices during the quarter. U.S. benchmark crude averaged $92.45 per barrel between April and June, up 27% from the previous quarter, as concerns over supply disruptions intensified following the escalation of the U.S.-Iran conflict.

The conflict has heightened fears over shipping through the Strait of Hormuz, a vital artery for global crude exports, while attacks in the Red Sea have complicated alternative export routes.

Higher crude prices generally translate into stronger cash flows for oil producers, while tighter supplies of refined fuels such as gasoline and diesel have also boosted refining margins.

Production Reaches Record Levels

Chevron reported record U.S. oil and gas production of approximately 2 million barrels per day, supported by higher exports as disruptions in the Middle East tightened global supplies.

Worldwide production increased to 4 million barrels per day, up from 3.4 million barrels per day a year earlier, representing roughly 20% growth.

Exxon also achieved record operational performance.

The company said upstream production reached its highest level in more than two decades, excluding disruptions related to the Middle East.

Production in the Permian Basin reached a record high, while total global output rose to 4.5 million barrels per day.

The strong production growth reflects years of investment in low-cost shale assets and large offshore developments that continue to generate higher volumes even as global supply risks increase.

Refining Becomes A Major Profit Driver

The surge in fuel prices significantly improved refining profitability for both companies. Chevron’s refining business generated $4.9 billion in earnings, more than six times the $737 million reported a year earlier, as gasoline and diesel prices climbed following supply disruptions linked to the Middle East conflict.

Exxon’s refining segment reported $5.5 billion in earnings, marking a dramatic turnaround from a $1.3 billion loss in the first quarter.

Compared with $1.4 billion in the same quarter last year, the improvement was driven by strong Gulf Coast refinery utilization and record diesel production.

The results reveal that integrated oil companies are benefiting from multiple parts of the energy value chain during periods of elevated prices. While upstream operations capture gains from higher crude prices, refining businesses often earn stronger margins when fuel supplies tighten.

Upstream Earnings Strengthen

Chevron’s oil and gas production business generated $8.2 billion in earnings, compared with $2.7 billion a year earlier, reflecting the combined impact of higher oil prices and increased production.

Exxon’s upstream division earned $7.9 billion, up from $5.4 billion in the second quarter of 2025.

The strong upstream performance reaffirmed the sector’s sensitivity to crude price movements, particularly during periods of geopolitical disruption.

The results show that geopolitical developments have once again become a key driver of energy markets. Although ceasefire efforts have reduced immediate fears of a wider regional conflict, continued tensions involving Iran, the Strait of Hormuz and Red Sea shipping routes have kept risk premiums embedded in oil prices.

Cloud infrastructure provider Nscale has agreed to acquire AI software startup Anyscale in a deal reportedly valued at about $1.65 billion, marking another step in the rapid consolidation of the artificial intelligence infrastructure market as providers race to offer end-to-end platforms capable of powering increasingly complex AI workloads.

The acquisition, announced on Thursday, combines one of the emerging “neocloud” infrastructure providers with a fast-growing software company whose technology enables enterprises to deploy, manage and scale AI applications more efficiently across large clusters of graphics processing units (GPUs).

Bloomberg News, which first reported the transaction, said the deal is worth approximately $1.65 billion, citing a person familiar with the matter. Nscale declined to disclose the financial terms.

The transaction is expected to close in the second half of the year, subject to customary closing conditions.

San Francisco-based Anyscale develops software that allows organizations to orchestrate distributed AI workloads across thousands of computers simultaneously, helping customers optimize computing resources, improve reliability and reduce the cost of training and deploying large AI models.

Its technology is built around Ray, the open-source distributed computing framework originally developed at the University of California, Berkeley. Ray has become one of the industry’s most widely adopted frameworks for scaling machine learning, generative AI, data processing and reinforcement learning applications across large GPU clusters.

By integrating Anyscale’s software with its cloud infrastructure, Nscale aims to offer enterprises a more comprehensive AI platform that combines computing capacity with software tools needed to build, train and deploy advanced AI systems.

Infrastructure providers in the AI industry are increasingly seeking to differentiate themselves by offering fully integrated technology stacks rather than competing solely on access to GPUs.

“Customers increasingly want a single platform that provides not only compute capacity but also the software layer required to manage increasingly sophisticated AI workloads,” industry analysts have noted, as enterprises seek to improve efficiency while controlling rapidly rising AI infrastructure costs.

Nscale said Anyscale will continue operating under its existing brand while serving its current customer base. The startup’s approximately 200 employees across the United States, Europe and India will join Nscale following completion of the transaction.

Full-Stack AI Platforms Becoming The New Battleground

The acquisition reveals that competition in AI infrastructure has expanded well beyond simply providing access to Nvidia’s latest processors.

As companies invest billions of dollars in AI development, attention has shifted toward software that maximizes GPU utilization. AI accelerators are among the most expensive components of modern computing infrastructure, making efficient scheduling and workload management increasingly valuable.

Training and running large language models often requires thousands of GPUs operating in parallel across multiple data centers. Software platforms such as Anyscale’s help coordinate those distributed systems, ensuring workloads are balanced efficiently, minimizing idle hardware and reducing training times.

That capability has become increasingly important as enterprises seek stronger returns on massive AI investments.

Industry executives have repeatedly acknowledged that simply adding more GPUs does not guarantee better performance. The ability to orchestrate workloads efficiently is becoming a critical competitive advantage as AI clusters grow in size and complexity.

Neocloud Providers Challenge Traditional Hyperscalers

Founded in 2024, Nscale belongs to a new generation of cloud providers often referred to as “neoclouds.”

Unlike traditional hyperscale cloud providers, neocloud companies are purpose-built around AI infrastructure, combining ownership of data centers, GPU clusters, and software platforms into vertically integrated businesses optimized for machine learning workloads.

The model has attracted significant investor interest as demand for AI computing continues to outpace available capacity.

Nscale competes with specialist AI cloud providers including CoreWeave and Nebius Group, both of which have positioned themselves as alternatives to larger cloud providers such as Amazon Web Services, Microsoft Azure and Google Cloud for customers requiring dedicated AI infrastructure.

Rather than serving general-purpose enterprise computing, these companies focus on high-performance AI training, inference and model deployment, enabling customers to access large-scale GPU resources without building their own infrastructure.

The acquisition of Anyscale strengthens Nscale’s competitive position by adding a software layer that complements its infrastructure business, allowing it to compete more effectively for enterprise AI customers seeking integrated solutions.

The transaction is the latest example of consolidation sweeping through the AI infrastructure sector as providers seek to broaden their capabilities amid surging enterprise demand.

Rather than purchasing individual AI tools from multiple vendors, many organizations now prefer integrated platforms that combine computing infrastructure, orchestration software, model deployment and lifecycle management under a single provider. That trend has encouraged infrastructure companies to acquire software developers while software companies pursue partnerships with cloud providers to deliver more comprehensive AI ecosystems.

Anyscale said in a blog post that revenue grew 70% sequentially in its most recent quarter, highlighting the rapid growth in demand for software that helps enterprises deploy AI applications at scale.

The global AI infrastructure market has expanded rapidly over the past two years as companies accelerate investment in generative AI. While Nvidia has dominated the market for AI chips, a broader ecosystem has emerged around cloud infrastructure, data centers, networking, orchestration software and developer platforms needed to build and operate advanced AI models.

This has fueled the rise of specialist AI cloud providers, or “neoclouds,” including Nscale, CoreWeave and Nebius, which compete by offering dedicated GPU infrastructure optimized for AI workloads.

At the same time, software platforms such as Anyscale have become increasingly important because they enable organizations to efficiently manage distributed AI computing across thousands of GPUs.