Venture capitalists have significantly increased their investment in AI startups, dedicating over half a trillion dollars to the sector over the past five years. However, the more promising investment opportunity might now be in energy, as per a report by Sightline Climate. Researchers indicated that potentially up to 50% of announced data center projects could face delays, primarily due to power access issues.
Sightline is monitoring 190 gigawatts of data centers, with only 5 gigawatts currently under construction. Last year, around 6 gigawatts of data center projects in Sightline’s database came online, but about 36% experienced timeline slippage in 2025. These delays could potentially impact large enterprises and other companies utilizing AI.
This supply-demand challenge presents an opportunity for investors. Major tech companies like Google and Meta are heavily investing in solar, wind, and nuclear projects, and supporting innovations like Form Energy’s 100-hour battery through direct investments and utility collaborations.
Numerous startups are tackling the power dilemma. Amperesand, DG Matrix, and Heron Power are working on new power conversion technologies, while Camus, GridBeyond, and Texture are developing software for managing electron flow.
Power supply remains a major constraint for data centers, exacerbated by AI’s anticipated 175% increase in power consumption by 2030, according to Goldman Sachs. The grid shortages are unprecedented and have been driving up electricity prices, prompting many tech firms to seek alternative power solutions. The Trump administration is encouraging tech companies to develop their own power sources or accept higher rates.
Tech giants like Amazon, Google, and Oracle are reducing their grid dependency. Some data centers are being planned to use on-site power or a hybrid of on-site and grid power. The biggest data centers are leading this shift, with less than a quarter of projects opting for on-site or hybrid methods, accounting for 44% of total capacity.
Power generation equipment shortages and an outdated grid have paved the way for alternative energy sources. Google’s recent initiative in Minnesota illustrates a strategy to address the problem by combining wind and solar with a massive battery from Form Energy. Additionally, Google collaborated with Xcel Energy to create a new rate structure promoting new technologies in the utility’s planning.
Grid-scale batteries are positioned to significantly impact the power market, with the US expected to have nearly 65 gigawatts of battery storage capacity by year-end, as reported by the US Energy Information Administration. Form Energy, like many others, is seeking to harness this momentum by raising a $500 million round before a potential IPO.
Energy supplies represent only a part of the challenge. The power, once on the grid or in the data center, requires management, primarily by transformers. Most transformers use heavy iron technology over 140 years old, reliable yet bulky with increasing data center power needs. Solid-state transformer startups are gaining investor attention, exploring silicon-based power electronics as an alternative to the archaic iron-and-copper technology. While more costly, they promise flexibility to replace multiple data center components, potentially making them competitive.
Overall, investments in battery and transformer companies remain smaller than the large funding rounds in AI, but they present more manageable opportunities for investors. As the global demand for electrification grows, investing in power solutions may provide a safeguard against a potential AI downturn. Perhaps the best AI investment might not be in AI directly.