The consequences of the US midterm elections

IISS | Paul Fraioli, Benjamin Rhode, Dana Allin, Jonathan Caverley, Jonathan Stevenson

American voters delivered a decisive defeat to President Donald Trump’s Republican Party in the 3 November 2026 midterm congressional elections, handing Democrats control of both the Senate and House of Representatives. Exit polling identified corruption, cost of living, and the unpopular war against Iran as the primary drivers of voter discontent.

Democratic majorities seated from 3 January 2027 to January 2029 empower congressional leaders to block judicial appointments, scrutinise corporate executive deals, and leverage federal appropriations against White House initiatives. Defeat reshaped domestic energy and technology trajectories. Voters approved restrictions on artificial intelligence data-centre development across approximately 41 ballot measures nationwide, dampening capital investment expectations for high-performance computing infrastructure. High gasoline and diesel prices stemming from Middle East operations reinforced public opposition. Restricted by legislative paralysis domestically, Trump will likely rely on direct executive commands to drive trade actions, alliance management, and personal projects until January 2029.

Comment

State-level moratoriums on AI data-centre construction, such as the restrictions enacted in Texas, disrupt the dual-use computing pipeline essential for military algorithm training. Power grid interconnections managed by regional transmission organisations face severe operational strains from hyper-scale facilities. Legislative friction directly slows the deployment of sovereign high-performance computing clusters needed for advanced signal processing.

The resulting capital pullbacks shift compute-intensive workload expansion back onto federal infrastructure. Private sector infrastructure bottlenecks force defence contractors to compete for capacity on established platforms like Oak Ridge National Laboratory's Frontier supercomputer. Legislative constraints in key jurisdictions effectively delay the transition of commercial AI models onto dedicated systems within the Joint Warfighting Cloud Capability.

Strategic Question for Discussion

If state-level moratoriums like those in Texas restrict commercial data-centre expansion, how will the Department of Defense reconcile its reliance on commercial cloud platforms with the secure processing demands of the Joint Warfighting Cloud Capability?

The pattern of state-level regulatory pushback indicates that federal defense entities will increasingly need to insulate national security computing needs from civilian power grid controversies. My assessment is that the Pentagon will favor sovereign, federally owned installations over commercial leases to protect key algorithm training pipelines. This shift likely increases near-term capital expenditure for defense microelectronics while insulating core command systems from local political volatility.

Share your assessment in the comments below.

Read the original article