The Geopolitics of AI: How States Compete for Ultimate Tech Leadership
- hoffmann58
- 1. Apr.
- 5 Min. Lesezeit
Executive Summary
The United States under Trump has made AI dominance a top priority, repealing AI safety rules and adding 70 Chinese entities on an export blacklist for AI chips.
The EU has eventually started to shift away from its overt focus on AI regulation towards a competitiveness agenda to spur innovation within the Single Market.
China experiences an AI boom after DeepSeek’s moment with the Hang Seng benchmark index for tech stocks surging by more than 40% since mid-January.
Implications for International Business
The race to lead in AI provides opportunities, such as industries reaping productivity gains from the adoption of AI systems in their operations and benefiting from public investments that seek to drive innovation at home.
At the same time, complex and fragmented national AI regulations can lead to legal risks while export controls on advanced AI chips are likely to restrict the flow of technological goods.
State of Play
AI as a determinant of national power
Recent developments have demonstrated the fundamental shift that AI is causing on the global security and economic landscape, from AI-enabled drones in Ukraine to AI innovations awarded two Nobel Prizes in 2024. As a result, some governments aim to achieve supremacy over AI development. The United States has outpaced other countries, currently developing 70 percent of the world’s notable AI models. China is catching up, as indicated by the release of DeepSeek’s R1 model. In addition, the EU aims to become an AI continent by simplifying its tech rulebook and setting up AI factories that provide computing resources to European innovators. The EU’s push for AI innovation has been partly prompted by technological breakthroughs in the United States and China, as well as by the lack of European alternatives. The UAE and Saudi Arabia have also identified AI as a key driver of their economic diversification from hydrocarbons, with Abu Dhabi having launched an investment firm with a target size of $100 billion to push for AI dominance.
Moreover, these developments have provided a geopolitical role for major corporations. Google has lifted its ban on using AI for developing weapons, while Meta has announced $65 billion investments on AI this year, with the aim to build the world’s largest AI data center. With these decisions, firms aim to become strategic actors themselves in the AI race. This tech-driven world could provide significant benefits to other international businesses too. Across the banking industry, the technology could deliver up to $340 billion annually, while in retail the impact could reach up to $660 billion. However, companies will need to navigate through the geoeconomic risks of the AI race, such as export controls and complex regulations.
State of Play
The geopolitical implications of AI and technology
The military use of AI-enabled weapons is growing. Ukraine’s military has deployed AI-equipped drones to strike Russian critical infrastructure. American AI systems have been used to identify airstrike targets in Syria and Yemen. Such rapid AI advancements could lead to further breakthroughs and create geopolitical tensions. For one, AI lowers the barrier for malign actors, including terrorist groups, to develop and produce chemical or biological threats by providing detailed instructions for assembling viruses. For another, AI applications pose severe risks for governments as they can tip the military balance in favor of adversaries. To mitigate this scenario, governments aim to spur innovation and maintain their position in the cutting-edge of technological development. The US Stargate project aims to achieve that by investing $500 billion over the next four years towards building new AI data centers for OpenAI. In the EU, President von der Leyen announced €200 billion of public and private investments to boost the continent’s AI competitiveness. While it is uncertain if these investments will yield concrete results, governments clearly aim to develop and secure sovereign AI capabilities that will provide security and economic benefits.
In addition to AI development efforts, governments often aim to impede the progress of their adversaries. Washington has imposed a series of economic restrictions on China, such as blocking Chinese providers from direct access to the advanced semiconductors required for frontier AI development. The Netherlands and Japan, who hold a strategic position in the global semiconductors value chain, have aligned with these restrictions, directly affecting firms such as ASML, which sourced 26 percent of its sales from China in 2023. In January 2025, the United States not only expanded export controls to more countries but also introduced further restrictions on AI model weights – the parameters containing a model’s capabilities. This is likely to impact US providers like Nvidia and AMD which are faced with significant revenue threats. Besides absolute restrictions, a series of countries face limits on the number of AI chips that can be deployed within their jurisdictions. For example, countries like the UAE, India and Mexico cannot import more than 7 percent of Nvidia’s or AMD’s AI chips. These policies have amounted to a new form of technological protectionism, in which America makes use of economic statecraft and its diplomatic relationships to preserve its advantage. DeepSeek’s CEO himself has described these restrictions as the most significant challenge that his company faces.
Geoeconomic risks and opportunities in the AI race
The integration of AI into the geopolitical calculus of nation states generates both political and financial risks. Jurisdictions have adopted regulatory frameworks to implement guardrails and ensure the safe deployment of AI systems, such as the EU’s AI Act, which addresses AI systems with a potentially adverse impact on people’s safety or fundamental rights. For example, it bans systems that exploit vulnerabilities of individuals or use manipulation, while strictly regulating systems deployed in sensitive areas such as education, law enforcement and critical infrastructure. Failure to comply can lead to fines of up to 7 percent of a company’s global annual turnover or €35 million, whichever is higher. This has prompted companies such as Meta, OpenAI and Apple to delay the launch of some AI products in the EU market. At the same time, other jurisdictions such as the United Kingdom and the United States have opted for differentiated regulatory frameworks. In the latter alone, the number of pending AI-related bills at both the federal and state level has risen to 781 in 2025. Such regulatory divergences can increase the cost of compliance and impose legal uncertainty on international businesses. In Colorado for example, developers need to conduct AI Impact Assessments which can differ significantly in the types of AI systems that are covered and the required obligations in comparison to the EU’s conformity assessment. Finally, countries and regions with a low level of AI preparedness, including limited connectivity infrastructure and AI talent, will not be able to reap the military and economic benefits of AI and risk falling behind.
There are also significant opportunities that the AI race presents for companies. Firstly, the mobilization of public investments creates financial opportunities for experimentation and the adoption of AI within traditional industries. The EU plans to dedicate €20 billion of its AI investment portfolio to a European AI Fund which will direct resources for AI research, innovation and diffusion. Secondly, governments are ushering a simplification agenda to be able to compete with their rivals. London aims to establish ‘AI Growth Zones’ to streamline the planning approval process for AI data centers and facilitate the provision of clean energy sources. Currently, businesses in the EU must navigate a complex patchwork of 27 distinct corporate regimes. In 2026, the EU plans to introduce a ‘28th regime’ to offer an alternative harmonized legal system that firms can opt into and benefit from harmonized corporate requirements. Thirdly, businesses stand to benefit from the competition between AI providers. McKinsey reports that 42 percent of organizations see cost reduction from AI implementation and 59 percent experience revenue increases. Therefore, the race between AI providers to develop better and cheaper products can have direct positive impacts for both large corporations and small businesses.