Artificial intelligence is starting to reshape how oil and gas companies find, drill, and produce fuel. A new report from McKinsey & Co. lays out just how much money is on the table.
The consulting firm published its findings on August 25. It looked at more than 550 individual AI use cases across the upstream oil and gas life cycle.
McKinsey estimates AI could generate about $65 billion a year in value using tools already available today. That number climbs to $125 billion as companies adopt proven technology more widely.
At full potential, with autonomous systems running across operations, the yearly value could reach $230 billion. These figures already subtract more than $30 billion in yearly costs to run the AI systems.
Where the Value Is Concentrated
The opportunity is not spread evenly. McKinsey found that the top 10 AI use cases drive close to half of the total value.
The top 20 use cases account for about two thirds. The top 60 make up roughly 95 percent of the value identified in the study.
Most of that value sits in production optimization and drilling. AI can help run pumps, gas lift systems, and production networks more efficiently.
In drilling, AI can help plan wells, choose equipment, and spot problems early. It can also help place wells more accurately within a reservoir.
McKinsey says these tools could shrink the time needed to explore and develop a field. Work that once took months or years could take days or weeks instead.
The report also points to a separate gain. Better AI driven exploration could add more than $35 billion a year in balance sheet value through new reserves.
Companies Already Testing AI Tools
Some of this work is already underway. SLB and Vår Energi are building shared AI tools for well planning on the Norwegian Continental Shelf.
The two companies say the goal is to cut the time from a discovery to first oil from months to days.
Baker Hughes and Expand Energy have struck a multiyear deal as well. They plan to roll out an AI powered production tool called Leucipa across thousands of natural gas wells.
McKinsey says the biggest barrier to scaling AI is not really the technology itself. Many companies are focused on the wrong problem, the report argues.
Incomplete data, older systems, and weak connectivity are real issues. But the report says AI only creates value when it actually changes how decisions get made.
Fewer than 20 percent of companies currently track how much value their AI tools generate, according to the report.
There is also a money question buried in the shift. Oilfield service companies could lose billable work as AI makes drilling and production faster and more reliable.
McKinsey estimates that up to $60 billion in oilfield services revenue could be exposed at full AI potential. At current deployment levels, that figure is closer to $17 billion.
Some companies are adjusting their contracts to match. Helmerich & Payne now uses performance based deals, earning bonuses for beating drilling speed targets instead of billing by the hour.
Other companies are testing similar setups, including software subscriptions and shared savings arrangements. McKinsey says these models could become more common as AI tools prove themselves in the field.
