The global energy sector is facing an unprecedented paradox. On one hand, the cost of solar panels is expected to drop another 30 percent over the next decade, paving the way for photovoltaic technology to dominate the world's energy mix by 2035. On the other hand, the relentless hunger for energy from AI data centers is creating such massive and constant base load demand that it will extend the life of fossil fuels, especially natural gas and coal.
According to an analysis published by TechCrunch today, this dual dynamic represents one of the most complex challenges for the green transition. Solar photovoltaics will not only become the cheapest source of electricity but also the one with the largest installed capacity globally, surpassing coal and gas. However, the intermittent nature of solar power requires large-scale storage solutions, and until batteries can cover nightly and seasonal needs, AI data centers will continue to rely on gas or coal plants to guarantee 24/7 uptime.
The Plummeting Cost of Photovoltaics and the Clean Energy Era
Advances in solar cell manufacturing, the adoption of perovskite materials, and economies of scale are driving the cost per kilowatt-hour of solar to unprecedented levels. Forecasts indicate a further 30 percent reduction by 2035, making solar competitive without subsidies in nearly every region. By that date, solar could account for over 40 percent of new electricity generation capacity installed annually. This is a victory for the climate, but overall electricity demand is growing at an unexpected rate, driven specifically by the expansion of AI infrastructure.
The Insatiable Energy Appetite of Artificial Intelligence
Data centers that train and run large language models, neural networks, and real-time inference systems consume enormous amounts of electricity. A single cluster of GPUs can draw as much power as a small town. While companies like Google, Microsoft, and Amazon are investing in renewable energy to power their facilities, the operational reality requires a continuous supply that solar alone cannot guarantee. AI data centers are therefore signing power purchase agreements that include combined-cycle gas plants, creating a parallel market that extends the economic life of fossil fuel infrastructure.
This trend is particularly evident in the United States, where environmental regulations are less stringent and natural gas production is abundant. Furthermore, the growing demand for electric vehicles, as illustrated by the launch of the Volvo EX60 with a starting price of $58,400, adds additional pressure on the electrical grid. Although mass adoption of EVs is positive for local emissions, the increase in consumption requires stable baseload generation.
Implications for the Energy Transition
Experts warn that the world cannot afford to slow down decarbonization. If AI data centers keep coal and gas plants alive for another twenty or thirty years, the global carbon budget could be compromised. The solution lies in massive investment in long-duration energy storage, such as flow batteries, green hydrogen, and thermal storage. Only when solar energy can be stored for days and weeks at competitive costs can data centers fully abandon fossil fuels.
Looking Ahead
The energy landscape of 2035 will be marked by a paradox: solar will dominate energy production, but fossil fuels will not yet be dead. Artificial intelligence, paradoxically, could become both the driver of renewable energy demand and the lifeline for the fossil fuel industry. For further context on this technology transformation, refer to the Wikipedia page on solar power. The challenge for the coming years is to make the energy transition not just dominant but also definitive.
Sponsored Protocol
