Predictions, Validations

Major Validations

Three results anchor the program's public record.

The thirty-nine-day collision call. MindCast's November 16, 2025 simulation forecast that federal control over AI data center grid connections would trigger state resistance grounded in the Federal Power Act, litigation warnings from former regulators, and state counter-coordination. The Wall Street Journal's December 26 reporting documented every element — six of six institutional dynamics confirmed or actively materializing.

The state-response mechanism. MindCast forecast that states would answer federal acceleration by protecting their own ratepayers rather than blocking projects outright. Florida delivered first, barring utilities from charging residents for data center development, and California, Ohio, and Utah followed with rules requiring developers to pay their own energy costs — the exact direction MindCast's model identifies as where the field settles.

The downstream-constraint thesis. MindCast's controlling claim — federal permitting acceleration increases the share of siting outcomes decided at state and local level — now carries a twenty-seven-state legislative field as its evidence base.

Major Outstanding Predictions

The forward book below is live. Each entry carries a confidence band and resolves against municipal dockets, commission proceedings, statutes, and corporate disclosures through July 2028.

• By 2028, community benefit negotiations rank among the top three obstacles to building large data centers — alongside power and land (88–93%).
• A water, rate, or subsidy controversy involving one data center operator gets cited by opponents against unrelated companies in other states (85–92%).
• Rate guarantees, water commitments, and cost protections replace jobs-and-taxes promises as the headline concession developers offer in contested sitings (76–84%).
• More states and utility commissions adopt standardized rules making data centers pay their own grid costs — deposits, minimum-payment terms, protection against stranded infrastructure (80–88%).
• Most likely endgame through July 2028: standardized state cost rules plus project-specific local benefit deals — rather than moratoriums or federal preemption (~52%).

Core Publications

The publications below carry the program's full analysis. Each summary states the paper's controlling contribution, so readers can choose a starting point without reading all of them.

The Federal-State AI Infrastructure Collision — In November 2025 MindCast simulated what federal control over AI data center grid connections would trigger: state resistance under the Federal Power Act, litigation warnings, coordinated pushback. The Wall Street Journal documented those dynamics thirty-nine days later. The paper grades the forecast and maps three scenarios through 2028, with partial federalization — federal control over process, state leverage over place and cost — as the most likely outcome.

The Power Stack — How Energy Infrastructure Became the New AI Battleground — Energy, not chips, is becoming the bottleneck that decides who wins in AI. The paper maps how transformer shortages, grid-connection queues, and long-term power lockups concentrate market power — and why antitrust scrutiny will follow the bottleneck upstream from AI applications to energy and grid access.

The AI Infrastructure Energy Opportunity Landscape — Investment is pouring into data centers, chips, and models while the real bottlenecks — transformers, transmission, advanced generation, grid software — stay comparatively underfunded. The paper maps the gap and argues that removing bottlenecks, not capturing them, is the investment position that survives both the market and the enforcement cycle.

The Two-Ledger Data Center Bargain — Data centers keep losing county votes the tax math says they should win. The paper explains why — residents weigh feared losses like water and rate increases roughly twice as heavily as promised benefits — and shows that developers who deliver enforceable protections before opposition organizes win approvals that larger benefit packages lose. A falsifiable prediction register through 2028 and a twelve-move operating playbook convert the model into practice.

The same architecture can be purpose-built for a campaign your organization is tracking or party to.