AI, Energy, and Geopolitics

A synthesis report by FP Analytics in partnership with the World Governments Summit

The global race for AI dominance has intensified with the arrival of DeepSeek, a China-based AI model that rivals models from the United States at a fraction of the cost. The introduction of DeepSeek and other more cost-effective AI models could accelerate AI adoption across economies, driving the need for resources such as energy and water to power AI globally. For instance, data centers supporting AI operations are projected to consume 1,580 terawatt-hours per year of electricity by 2034, comparable to the electricity consumption of India.

Against the backdrop of these rapidly evolving advancements in AI, in February 2025, FP Analytics and the World Governments Summit held an in-person, interactive simulation in Dubai, United Arab Emirates, bringing together practitioners, policymakers, and experts to explore the intersection of AI, energy, and geopolitics. Participants grappled with how to balance AI innovation and growth with resilient energy systems and sustainable resource management that will be central to future economic growth, competitiveness, and security. As part of the simulation, participants were assigned diverse roles across the public and private sectors, including the United States, China, India, and Brazil; the European Union (EU); the Gulf Cooperation Council (GCC); the International Organization for Standardization (ISO); the International Renewable Energy Agency (IRENA); the World Bank; and fictional coalitions representing the tech industry, the energy industry, environmental advocates, and consumer advocates. They navigated three “moves” or fictional scenarios, each introducing security, socioeconomic, and humanitarian risks around the pursuit of sustainable AI adoption. This synthesis report distills key takeaways from the dynamic and immersive simulation that highlighted the need for cross-sectoral partnerships to address cascading risks and harness opportunities at the nexus of AI and energy.

Move 1: The Race to Build AI Opportunity Clusters (2027)

In Move 1, participants saw governments and the private sector scrambling to expand data center clusters in strategic locations around the world. Protective trade restrictions and retaliatory measures threatened global supply chains, particularly for high-tech goods and critical minerals and materials. Moreover, tensions over land use, high water consumption, and opposition from local communities resistant to large-scale infrastructure development complicated efforts to launch AIOCs. Sustainably expediting AIOC development in major hubs in MENA, North America, Latin America, and Asia required striking a balance between meeting short-term energy needs and boosting long-term socio-economic development.

In an increasingly multipolar world with shifting trade alliances, participants identified new opportunities for collaboration among countries that could benefit from each other’s expertise and resources. The group raised possibilities of setting aside geopolitical issues, such as border tensions between China and India, to jointly develop power infrastructure and exchange AI technologies and resources. They emphasized diversifying sources of critical minerals and investing in green AIOCs particularly in lower- and middle-income countries (LMICs) to drive those economies, provide capital and jobs, and bring about regional stability. Participants also discussed “big tech’s” role as a primary driver of power demand, highlighting the need to engage with industry leaders and demonstrate business benefits of establishing safety standards and integrating environmental and local development considerations into AIOC expansion. Multilateral institutions can help unlock capital for green infrastructure projects by facilitating public-private financing, but these efforts must be underpinned by clear, agreed-upon standards for AI safety and sustainability.

Move 2: Cybersecurity Crisis Destabilizes AIOC Development (2029)

In Move 2, participants faced a multipronged cyberattack on a nuclear facility in Busan, South Korea, exposing system and workforce vulnerabilities in AI-integrated energy systems. Governments apply a critical infrastructure designation to the AIOCs, raising the stakes of international partnerships and highlighting uncertainties regarding cluster governance and management. As the attack’s origins remained unclear, the crisis triggered a range of accusations, increasing tensions among major powers, sowing misinformation, and further eroding public trust.

This move highlighted the dual-use nature of advanced technologies, which can drive industrial progress while also being weaponized for malicious purposes, fueling mistrust and complicating international cooperation. In response, participants explored the role of middle powers and nonaligned territories with economic and security ties to competing major powers as potential hosts for AIOCs, fostering AI and tech development in more stable environments. To strengthen cyber resilience, they discussed issues surrounding data delocalization, and the potential to split and distribute data across multiple locations – potentially even in space – rather than confining the data to one physical location. This approach and “debunking myths” around data localization could help to reduce the risk of cyberattacks compromising entire ICT systems. In addition, enhancing AI and digital skills across workforces were noted as being vital to strengthening cyber resilience. However, for this to happen, AI needs to become more affordable and use cases expanded to enable wider adoption. Given the evolving nature of cyber threats, participants stressed the need for continuous public-private collaboration to stay ahead of security risks and ensure AI and energy infrastructure safety.  

Move 3: Hydropower Collapse Catalyzes Global Summit on AIOC Governance (2032)

In Move 3, a hydropower dam that was part of an AIOC in Rio Grande do Sul, Brazil, collapsed, resulting in humanitarian casualties and severe economic disruption. Global infrastructure developers and government officials came under intense scrutiny for truncated project development timelines and lax safety standards. The disaster’s aftermath reverberated globally, prompting an urgent call for action. A high-level summit in Brazil convened leaders from governments, industry, academia, and civil society to re-assess AI/Energy cluster sustainability and security, as well as means of safeguarding buildouts at scale.

The discussion under the third move focused on the societal and environmental impacts of AI and energy development, tying together reflections from all three moves as the simulation came to an end. Participants raised the persistent challenge of a shortage of financing options that support the long-term returns of climate-resilient and low-emissions projects, as current financing options favor quicker, shorter-term returns. Climate-resilient projects continue to face funding and execution difficulties, especially in LMICs. One key question that emerged was how stakeholders could mobilize private-sector capital and sovereign wealth funds to finance green projects effectively. Regarding energy and geopolitics, a participant questioned the role of cryptocurrency in energy and geopolitics and its involvement in funding cyber risks and global influence operations. In addition, the discussion explored the possible role of China in fulfilling the financial gap in global climate and development funding, as the United States reduces its foreign aid. Participants reflected on the broader challenge of how global institutions can keep pace with innovation, particularly in the field of AI and cybersecurity, to better support the governance and funding of infrastructure projects.

Key Takeaways

Nonaligned nations are poised to solicit AIOC-related investment from competing major powers. Countries could seek partnerships with the United States and China in an effort to benefit from the flow of investments across the AI industry. Middle powers could also deepen cooperation among themselves to advance AI and power infrastructure buildout, potentially mitigating some geopolitical tensions in favor of economic cooperation.

As traditional sources of foreign aid diminish, new sources of finance can fill the vacuum in low- and middle-income countries (LMICs). Countries and regional blocs such as China and the Gulf Cooperation Council can play a greater role in financing AI and energy infrastructure projects and facilitating supply chains in LMICs. Meanwhile, the private sector and sovereign wealth funds could also help unlock the flow of capital into climate-resilient and low-carbon projects by valuing the long-term economic benefits of these projects.

Major tech companies remain primary drivers of AI innovation and resource consumption. Their significant influence reinforces the need for deeper public engagement with the industry to ensure accountability and demonstrate the business case for integrating safety and sustainability into AI and energy ecosystems.

AI and energy investments in LMICs could contribute to regional economic development and geopolitical stability. Workforce development, particularly in countries with young populations, is critical, as neglecting this area could lead to broader socioeconomic repercussions and regional instability. Investments in these areas also lead to diversified access to resources, which are key to minimize the supply chain vulnerabilities exacerbated by geopolitical risks.

Cyber threats continue to evolve, requiring proactive solutions. Key to combatting cybersecurity risks is public-private collaboration to stay ahead of security threats and ensure AI and energy infrastructure safety. Splitting and distributing data across multiple locations (data delocalization), rather than confining the data to one physical location, can help strengthen cyber resilience.

Looking Ahead

As in any situation where novel technologies are beginning to reshape industries and societies, participants in this simulation on AI, energy, and geopolitics raised important questions while brainstorming strategies and solutions to risks and challenges. In navigating the intersections of these key themes, dialogue and negotiations among diverse sectors will be critical to balance competing agendas and interests, and to arrive at outcomes that drive growth, safeguard safety, prioritize sustainability, and expand access. At the recently concluded Paris AI Action Summit in February 2025, the Statement on Inclusive and Sustainable Artificial Intelligence for People and the Planet emerged as a key framework for AI governance from the summit, but the notable absences of the United States and the United Kingdom as signatories to the AI framework indicate an emerging global impasse in AI governance. Without clear and enforceable standards, the world will be left unprepared for high-risk scenarios involving AI and energy systems that can create and exacerbate security, socioeconomic, and humanitarian challenges, and in turn carry political consequences. One key takeaway is the expanding role and influence of middle powers such as the UAE, Brazil, and India, and growing economies with young populations into the future. All of the actors in this scenario, from nation states to industry leaders to consumer advocates, will be shaping AI safety and sustainability policies regionally and globally. A number of high-profile summits are on the horizon, including the Global AI Summit on Africa, the 3rd UNESCO Forum on the Ethics of AI, and the AI For Good Global Summit 2025, providing venues for further discussions and much-needed collaboration on AI innovation, security, and sustainable resource management.

By Angeli Juani, Senior Policy and Quantitative Analyst, FP Analytics.

This synthesis report by FP Analytics, the independent research division of The FP Group, was produced with support from the World Governments Summit. FP Analytics retained control of the direction and findings of this report. Foreign Policy’s editorial team was not involved in the creation of this content.