A server rack is not what artificial intelligence looks like in the future. It has a little diesel and hot metal fragrance. The iconic shadows of rigs puncturing layers of shale and pumpjacks swaying slowly against a flat Texas sky are still visible if you drive west of Houston. However, officials at Baker Hughes, nestled away in glass office towers downtown, are more concerned with megawatts than drilling depth. Not the type that runs refineries, either.
The oil business is going through a difficult time. The price of crude has decreased. Tensions in geopolitics persist. Pressures from the climate are still increasing. Rivals such as Halliburton and SLB reported net income drops of 33% and 9%, respectively, in the second quarter of 2025. In light of this, Baker Hughes announced a 21% increase in net profits. Unexpectedly, data centers were the cause.
| Category | Details |
|---|---|
| Company | Baker Hughes |
| Headquarters | Houston, Texas |
| 2025 Q2 Net Income Change | +21% year-over-year |
| Competitors | Halliburton (-33%), SLB (-9%) |
| Emerging Growth Driver | Gas turbines & energy systems for data centers |
| Official Website |
Few people outside the sector may have understood how closely oilfield services firms are now linked to the AI revolution. Standing close to one of the new hyperscale data centers that are emerging on the outskirts of Dallas or Northern Virginia, however, makes the reasoning more obvious. These buildings are always humming, with transformers buzzing behind chain-link fences and fans forcing air across GPU racks. Their electrical consumption is astounding.
Workloads related to artificial intelligence, such as training big models and doing inference at scale, require consistent, dependable power. Intermittency is still an issue, but renewable energy is helpful. Downtime cannot be afforded by data centers. Therefore, they are using backup systems and natural gas turbines from businesses that used to concentrate nearly entirely on oilfields. Something ironic about that.
Oil service companies have spent decades honing their skills in handling intricate energy systems in harsh environments. They became experts in pressure control, compressors, and turbines. These same abilities are now being used to power digital infrastructure. It’s difficult to ignore how smoothly the turnaround is taking place.
Engineers who used to optimize drilling operations are now talking about modular power units for data campuses at Baker Hughes’ Houston headquarters. Energy management is still the fundamental discipline, even though the terminology changes slightly from barrels per day to megawatts per rack.
Oil firms have long been undergoing digital transformation. For decades, the industry has been gathering and evaluating enormous volumes of data, including production indicators, drilling logs, and seismic readings. AI and big data analytics have subtly changed efficiency and exploration. What’s different is that the foundation of artificial intelligence is now provided by the oil industry. This change is perceived as presenting both risk and opportunity.
On the one hand, since traditional oilfield activity is slowing down, the data center boom offers a new source of income. Globally, liquefied natural gas terminals are growing, particularly as Europe looks for options in the face of continued tensions with Russia. Data center requirements readily correlate with investments in gas infrastructure.
However, environmental inspection is becoming more intense. Activists wonder if using fossil fuels to power AI just moves emissions instead than lowering them. Green initiatives that call for more accurate air pollution measurements are being promoted by governments. Oil service companies must strike a careful balance between meeting their climate responsibilities and fostering digital growth. Whether this new position will change how the public views the oil business is still up in the air.
You might hear the hum of cooling systems being tested and cranes moving prefabricated modules into position if you’re standing close to a new data center that is currently being built. Soon, rows of servers within will handle billions of queries, create graphics, and train language models. Outside, fuel is silently converted into power by gas turbines. Two universes that are becoming more and more entangled.
Energy security appears to be a top priority for investors in the AI era. A parallel race to find reliable power sources has been sparked by the rush to construct larger models. Businesses with extensive experience in heavy infrastructure profit from such dynamic.
However, there is a warning behind this. The markets for gas and oil are cyclical. Even though AI demand is currently skyrocketing, it may level off. There are risks associated with overbuilding capacity.
It seems as though history is repeating itself as we watch this play out. Fossil fuels propelled the industrial era. Something more abstract and cleaner was promised by the digital era. And now that data centers are proliferating, the old energy titans are once again essential—not as antiques, but as facilitators.
The unexpected contribution of the oil business to the data center era isn’t particularly noteworthy. It doesn’t make headlines go viral. However, behind the scenes, balance sheets are adjusted, contracts are signed, and turbines spin.
The most sophisticated algorithms in the world might operate on code and silicon. However, many of them are currently driven by gas fires and pipes, which are reliable, useful, and very much alive in the Texas heat.