The 2026 midterm election results brought into sharp focus a question that once seemed remote from Maryland’s tech economy: Who will pay for the artificial intelligence boom, and who will collect the returns?
Affordability concerns helped drive the Democratic victory in both chambers of Congress. The subsequent market selloff, including a 5.5% decline in the Nasdaq on Nov. 5, reflected growing uncertainty about AI investment returns and potential restrictions on data-center expansion.
For Maryland, these national developments have a particularly immediate dimension: electricity costs.
The state’s Office of People’s Counsel has challenged how regional transmission costs are allocated. It estimates that Maryland customers could pay $1.6 billion over 10 years for infrastructure projects driven substantially by growing data-center demand, much of it outside the state.
The estimate is part of a contested regulatory dispute, but the underlying economic question is straightforward. Should Maryland households help finance AI infrastructure investments from which they may receive relatively little direct benefit?
The AI industry has one balance sheet. Society has another.
The industry measures investment, computing capacity, revenues, and shareholder returns. Societies must also account for electricity prices, public infrastructure, employment, competition and opportunities for local businesses. Corporate profitability and public benefits can reinforce each other. But they are not interchangeable.
For Maryland, this is more than a utility-rate argument.
Our researchers, clinicians, developers, and entrepreneurs have substantial opportunities to apply AI in healthcare, cybersecurity, logistics, education and financial services. The economic value will come less from counting servers than from translating technological capabilities into products, services, and businesses that succeed in the market.
That requires affordable access to computing capacity, skilled people and opportunities for smaller firms to compete. Indiscriminate restrictions on data-center development could raise computing costs and inadvertently strengthen the large incumbents that already control substantial infrastructure. But shifting new grid costs onto households is no solution either.
Maryland needs an approach grounded in net economic benefit:
- Regulators should identify who drives new electricity demand and allocate costs fairly
- Policymakers should evaluate the actual local returns from major investments, not just headline spending figures
- Economic development efforts should consider whether smaller businesses can participate meaningfully in the gains
Can sustained interaction with AI strengthen human judgment and creativity?
There is another potentially important return that conventional economic measures rarely capture: human capability.
Emerging research including my framework, human-AI cognitive co-evolution (HACC), explores whether sustained interaction with AI can strengthen human judgment, creativity and problem-solving capabilities rather than simply replace cognitive tasks. The distinction raises questions worth considering as employers and educational institutions integrate AI into their operations.
Productivity gains matter. But so does whether people develop lasting capabilities or become increasingly dependent on systems they use without fully understanding.
The midterms were not a referendum on AI. They did, however, reinforce a broader public expectation that economic progress must translate into tangible improvements in people’s lives.
Maryland should judge the AI revolution by the businesses that grow, the jobs and capabilities it creates, the costs residents bear, and the benefits that remain here.
The real test is not how much AI infrastructure gets built. It is how much of the AI dividend reaches Maryland.
