Geopolitics Complicates AI M&A Landscape

The Committee on Foreign Investment in the United States (CFIUS) now scrutinizes a broader range of transactions, extending its reach beyond controlling stakes to include non-controlling investments in businesses dealing with critical technology, infrastructure, or sensitive personal data. This heightened focus is particularly intense for deals involving AI and machine learning, with CFIUS actively investigating transactions that were not formally notified. A "Reverse CFIUS" program, effective January 2, 2025, now restricts outbound U.S. investments into China, Hong Kong, and Macau. The regulations specifically target investments in companies involved with AI systems that have military or surveillance applications, as well as quantum computing and semiconductors. This policy reflects a bipartisan commitment to aggressive evaluation of both inbound and outbound investments, particularly concerning China. The European Union's AI Act classifies AI systems by risk level, imposing stringent requirements on "high-risk" applications concerning technical documentation, risk management, and quality control. These regulations add complexity to M&A due diligence, as non-compliance can significantly impact a target company's valuation and post-acquisition integration. Big Tech firms like Microsoft have also faced scrutiny for partnerships with smaller AI labs that fall short of traditional mergers but still concentrate market power. China has explicitly defined AI as a strategic technology to bolster its geopolitical and military strength, aiming for global dominance by 2030. This national strategy integrates state direction with aggressive private sector execution, creating a powerful ecosystem for AI development. The government actively supports domestic chip production and facilitates data sharing between agencies and private firms to accelerate AI progress. The acquisition landscape is shifting toward companies specializing in agentic AI, which are systems capable of autonomous action and decision-making. Valuing these early-stage companies is complex, as their primary assets are often intellectual property and research talent rather than revenue. Dealmakers are increasingly using agentic AI to automate due diligence, conduct real-time market analysis, and identify acquisition targets. U.S. export controls now cover a wide array of emerging technologies, including AI, machine learning, and advanced computing. These regulations restrict the sale of both hardware, like advanced AI chips from Nvidia and AMD, and software to certain countries and entities, impacting the global AI supply chain. The Department of Commerce's Bureau of Industry and Security has recently increased enforcement, targeting efforts to evade these controls. A recent executive order directs federal agencies to challenge state-level AI regulations that are deemed "onerous" or that create barriers to U.S. leadership in AI. This initiative establishes a task force to litigate against state laws that could hinder innovation or unconstitutionally regulate interstate commerce. The order also links compliance to federal funding, pressuring states to align with a more uniform national standard for AI governance. In a significant move, the Trump administration has ordered federal agencies to cease using AI products from Anthropic, labeling the company a "supply chain risk" after it reportedly refused to grant the U.S. military unrestricted access to its models. This has led to the cancellation of over $200 million in federal contracts and prompted agencies like the State and Treasury departments to transition to competitors such as OpenAI.