The next big AI gap may be net energy access

In July 2025, much of Europe popped under one of the hardest burning of its storms in recent memory. In Spain and France, thermometers pushed 40 ° C day by day and the grid came under the weight of increasing demand. According to EmberDaily use of electricity increased by 14% during the worst heat. A few months earlier, a sweeping blackout left millions in Spain and Portugal without power. In this context, the AI ​​explosion with energy-efficient data centers and the ever-expanding calculation needs-does not only look like a miracle of progress. It looks like a match.

Each head about AI discoveries floats more than an invisible increase – an electricity and consumption of water that is quietly drawn to the edges of already strained systems. In accordance with International Energy Organization (IEA)Global electricity consumption by the data centers is projected to exceed 2030, with AI being the only largest driver. But not all countries can keep up. And not all power is created equal.

The shift of energy behind the development of AI

THE Ai boom It remodes more of the technology industry. It changes the way the world’s infrastructure appears. Large -scale models, the foundation of today’s genetic developed AIs, place huge demands on electricity and require huge amounts of water to cool. For the context, researchers in MIT Say that “the computational power required for the training of AI genetic models may require an amazing amount of electricity, which leads to increased carbon dioxide emissions”, adding that “a large water is required to cool the material, which can accelerate municipal water supplies”.

In this context, clean and affordable energy is no longer a footnote. It is a fundamental AI readiness lever.

“Electricity will determine the AI ​​landscape just as oil was once defined,” said Kenso Trabing, its founder and chief executive Morphware ai. “Nations that can offer clean, abundant and affordable power will become natural magnets for AI infrastructure.”

Morphware is part of a new generation of infrastructure companies planning around the first day. Instead of excavating clean energy in heavy coal grids, the company created its basic activities in Paraguay-hosting the ITAIPU Dam, one of the largest sources of hydroelectric power in the world. This location, according to Trabing, was not just an economic decision, but also reflected the company’s belief that the energy strategy should be fundamental, not a second thought.

“It gave us access to low cost, renewable energy sources-which, for us, were critical not only financially but also strategically,” Trabing told me. “In the next decade, I expect that we will see a re -election where areas with surplus green energy – South America, the Middle East, the parts of Africa and some European hubs – begin to exit traditional technological centers such as Silicon Valley, just because they will not be able to finance.”

Beyond the emissions: Pure energy as a strategic ditch

While viability is public concern, Morphware’s approach emphasizes that renewable energy is also a private advantage. The execution of hydroelectric power gives the company a great cost advantage – especially compared to similar facilities in the US or Europe. It also protects them from the types of fossil fuel prices that can send the data center accounts.

“For us, renewable energy is more than cutting the broadcasts,” Trabing said. “Brings the stability and flexibility of prices – two things that matter when operating on a scale.”

There is also a growing appeal to businesses trying to reduce their Scope of 3 emissions. “As industries adopt the AI, more and more they want to align themselves with providers who can prove viability, not only to claim it,” Trabing added. “For Morphware, renewable energy is both a business advantage and a moral check.”

Still, building in underdeveloped geographical areas is not without its challenges. Access to specialized work, latent status issues and the uncertainty of politics remain real restrictions – but for some companies, trade is worth it.

These are not abstract benefits. A location such as Paraguay provides natural cooling from the proximity of hydroelectric power, which helps to alleviate the thermal water intensity management. And the ability to escalate the AI ​​infrastructure in areas that have not historically been considered as technological powerhouses opens a new vision of geographical decentralization.

Location as leverage

Morphware’s infrastructure now covers both Paraguay and Abu Dhabi-two sites that are not often combined with technological decks, but more and more relevant to a climate-limited world.

“Our decisions are always guided by two principles: plenty of pure energy and global connectivity,” Trabing said. “Paraguay gives us unparalleled access to renewable hydroelectric power to the ITAIPU Dam., Abu Dhabi, on the other hand, provides a strategic gateway between Europe, Asia and Africa.”

This reflects a broader shift: in a world where energy becomes the primary limitation of AI, the calculation will emigrate to wherever power is cheap, clean and politically stable. “Together, these sites reflect a strategy for building in energy rich in energy first, then connect these foundations to the broader AI ecosystem.”

But taking there was not easy, Trabing noted when I asked him about the challenges of the industry faced by the company. “We had to build infrastructure from scratch – roads, transformers, internet agreements – while bridging cultural and educational gaps,” Trabing said for the early days of Morphware in Paraguay. “The lesson for other builders is that emerging markets require patience and humility, but repayment is huge.”

Redesigning the global computing map

If pure energy becomes the decisive variable in AI infrastructure, then the map of world technology is to shift. “I predict a decentralization of the AI ​​infrastructure,” Trabing says. “Instead of everything concentrated in the US or China, we will see computational hubs spread in areas with energy surpluses.”

This vision has geopolitical consequences. “Politically, the energy will become part of the AI ​​strategy, with governments facing pure energy not only as a climate issue, but as a competitive necessity,” Trabing added. “Economically, the advantage will shift to nations that can export ‘calculation’ powered by pure energy, as is once exported by oil or production capacity.”

This framework was redefined as a purely technological race for weapons, but as an infrastructure and ecological. The real advantage may not be to whom it builds the fastest models – but to whom it can maintain them without destabilizing the network, planet or communities around them.

As global demand for calculation increases, the gap between Ai Haves and-nots can fall more and more according to its lines energy access. Those with abundant, accessible power will build. Those who cannot struggle to escalate, regardless of talent or ambition.

Morphware is not only in the review where the AI ​​infrastructure belongs. The companies in Iceland, Kenya and beyond betting are also betting on net power as the backbone of calculation. The actual shift in progress? This is not just who can build, but who can feed it, viable and on a scale. There may be the future of AI.