Silvia Weko: AI, Big Tech, and the Energy Transition

Dr. Silvia Weko

In May 2024, the Saint Pierre International Security Center (SPCIS) launched the “Global Tech Policy at the Forefront” series, featuring conversations with leading experts on the impact of emerging technologies—such as AI, blockchain, biometrics, and robotics—on global governance and public policy.

On April 8th 2025, we had the pleasure of interviewing Dr. Silvia Weko, a postdoctoral researcher at Friedrich-Alexander University Erlangen-Nürnberg (FAU). Dr. Weko joined FAU’s Chair of Sustainability Transition Policy in 2023, following her tenure at the Research Institute for Sustainability (RIFS). Dr. Weko’s research spans topics such as the geopolitics of green hydrogen, technology transfer, and the role of intangible assets in energy transitions. She has published extensively in leading journals, including Climate Policy, Energy Policy, and Review of Political Economy.

This interview delves into the findings from Dr. Weko’s research, focusing on how data and AI are transforming global energy systems, the risks of Big Tech dominance, and what a just and equitable energy transition could look like.

SPCIS：Good morning, Silvia. Thank you very much for joining us. I’m so excited about our conversation today. I came across your work through your papers and your LinkedIn profile. I know you’ve done extensive research on the great transitions, the geopolitics of energy, the socio-economic impacts of big tech, and issues of data monopolies—all of which align with our interests. To start, can you tell us about your current work and what excites you the most right now?

Silvia: Thank you for inviting me! It’s great to connect with others working on AI and AI governance. I noticed you’ve talked to Cecilia already—she was the second supervisor for my Ph.D., and we’re currently writing a paper together. It’s wonderful to see these kinds of networks forming.

Currently, I’m doing a postdoc at the Chair for Sustainability Transition Policy at FAU Erlangen-Nürnberg. Half of my time is dedicated to the NFDI4Energy project where our research group is collecting and analyzing data on policies for emissions reductions. The other half of my time I am teaching and doing research, including on big tech and energy geopolitics. One of my ongoing projects is a comparative analysis of Amazon, Google, and Microsoft and how they interact with energy companies. What I’ve found fascinating is how these companies are providing essential digital infrastructure for the energy transition. For example, they’re heavily involved in renewable energy projects, but at the same time, they’re attempting to serve oil and gas companies. Interestingly, for companies like Saudi Aramco—where oil extraction is relatively straightforward—AI tools aren’t as critical as they are for renewable energy systems. This contrast highlights AI’s unique role in clean energy systems versus traditional energy systems. That’s something I’m excited about right now.

SPCIS：That sounds fascinating! My next question relates to a key topic we’re exploring: AI’s environmental costs. We know that AI development consumes significant amounts of energy. What do you think are the environmental costs of AI development, and can AI also be a solution to the energy transition?

Silvia: That’s an excellent question. Even before the current AI hype surrounding tools like ChatGPT, there was significant discussion in energy circles about how AI and digitalization could transform energy systems. AI is critical for renewable energy systems because it helps solve problems related to weather prediction, electricity optimization, and storage management. For example, AI can automate decisions like when to charge electric vehicles or store energy, based on weather patterns and expected renewable energy availability.

However, on the flip side, AI also consumes a massive amount of energy. Running data centers, storing and processing vast amounts of data, and ensuring proper cooling for servers all require electricity—much of which still comes from fossil fuels. Some companies, especially in the U.S., are signing long-term contracts to use nuclear power for their data centers, but this doesn’t entirely solve the problem. Moreover, as demand for electricity grows due to electrification (e.g., electric vehicles, heating), the energy needs of AI and data centers can compete with other critical decarbonization efforts. So, while AI can help optimize energy systems, its environmental costs are a serious concern that we need to address.

SPCIS：That’s a very insightful view. Let’s shift to the geopolitics of energy and AI. How do you see energy consumption by AI influencing geopolitics, particularly in the context of U.S.-China competition?

Silvia: There are several layers to this. First, there’s the geoeconomic competition between tech firms. The U.S. supports its tech giants like Amazon, Google, and Microsoft as global champions, using their dominance as a form of infrastructural power. For example, U.S. tech firms provide cloud services and digital tools internationally, creating dependencies that reinforce U.S. influence.

China is also fostering its tech champions. While I’m less familiar with Chinese politics, I know Chinese cloud providers are expanding their services internationally. This creates a kind of “spheres of influence” dynamic, reminiscent of old-school geopolitics, where countries align themselves with specific tech ecosystems. States also support their energy firms in similar ways, which mirrors the dynamics of fossil fuel geopolitics. It’s fascinating to observe how these patterns are now extending to big tech and renewable energy.

SPCIS：That’s a compelling analysis. Since your research focuses on Europe, how do European countries position themselves in this competition?

 

Silvia:Europe’s position is evolving. I would say that initially, most European policymakers didn’t see the dependence on U.S. tech firms as problematic. They viewed it as efficient—after all, big tech companies are very good at what they do. But after political events like Trump’s election emphasized the risks of over-reliance on U.S. technology, I have started to see a shift among some policymakers. There has been some discussion about creating European alternatives, and whether Gaia-X could become something like this.  

However, progress has been slow, and many European companies and governments continue to rely on U.S. cloud providers for critical services. The challenge is that once you build your technology stack on a particular platform, switching becomes incredibly difficult, which further entrenches these monopolies.

SPCIS: Given this dominance, what measures can governments or international organizations take to counterbalance the power of big tech?

Silvia: One solution is to provide viable alternatives to big tech. This is expensive and often politically challenging, but in my opinion it makes the most sense because these digital infrastructures are natural monopolies. Another approach is regulation. For example, the EU has introduced rules to ensure data portability and interoperability, making it easier for companies to switch providers.

It’s also important to challenge the narrative that big tech dominance is inevitable. Policymakers often feel powerless because these companies are so entrenched, but history shows that monopolies can be broken. Public scrutiny and awareness are also essential. Most people don’t like the idea of big tech doing whatever it wants, so there’s public support for regulation.

Lastly, education is critical. Policymakers often don’t fully understand AI, cloud infrastructure, or their implications. Helping them grasp these concepts can empower them to make better decisions.

SPCIS: You’ve highlighted the role of data in big tech’s dominance. What kind of data governance is needed to support the energy transition?

Silvia: Data is crucial for the energy transition, but the challenge lies in how it’s managed. You need data from all parts of the energy system—generation (e.g., wind turbines), transmission and distribution infrastructure (e.g., grids), and consumption (e.g., electric vehicles, smart meters). However, this data is often siloed. For example, manufacturers like Tesla own the data from their vehicles and aren’t inclined to share it with grid operators.

One solution is to encourage data sharing between companies and public actors. Regulations could ensure that critical data is shared in a way that respects privacy while enabling better energy system management. Publicly owned data platforms could also help address this issue by providing shared access to key datasets.

SPCIS: Finally, since many of our audience members are in China, they’re very interested in the country’s ambitious carbon neutrality goals. What lessons can China learn from Europe?

Silvia: China’s approach to strategic planning is something Europe could learn from, particularly its ability to focus on long-term goals like electric vehicle adoption and renewable energy development. However, Europe’s emphasis on community participation could also offer valuable insights for China. For example, in Germany, communities often own renewable energy projects, which fosters public support and engagement.

One area where Europe struggles—and where both regions could improve—is moving beyond carbon pricing and emissions trading as the primary tools for decarbonization. Markets alone can’t solve infrastructure problems, such as the need for renewable energy grids or heating systems. Stronger regulations and public investment in infrastructure are essential.