An AI Data Center Boom, a $5.5 Trillion Skills Gap, and Two Bets on Faster Psychometrics

Four signals this week come from a business magazine, an AI-skills analytics firm, a peer-reviewed military study, and a pre-hire assessment vendor. None of them coordinated. But together they describe the same constraint from different angles: the AI economy is not short on capital or ambition, it is short on a fast, credible way to confirm who can actually do the work.

The AI data center boom's real bottleneck is people, not chips

Fortune reports that Alphabet, Microsoft, Meta, and Amazon are on track to spend roughly $700 billion building AI data centers in 2026 alone. But according to Randstad's CEO, the true limiting factor on AI's growth curve is not chips, energy, or capital. It is a shortage of specialized skilled labor. The industry needs an estimated 300,000 additional workers by 2030, spanning engineers, cybersecurity staff, electricians, HVAC technicians, and plumbers.

Veterans are named directly in the piece as an underused talent pool, and employers are already building practical, earn-and-learn pathways to reach them. Microsoft's Military Datacenter Pathway is one example: eight weeks total, seven of them online and one hands-on, paying a $20-an-hour stipend, leading straight into a Critical Environment Technician role. Salute has already trained more than 10,000 veterans for similar roles and is onboarding 10,000 more. A mainstream business outlet naming the AI boom's actual bottleneck as provable practical skill, with veterans positioned as the answer, is exactly the thesis PowerTechs is built around. (Source: https://finance.yahoo.com/technology/ai/articles/tech-giants-spending-700-billion-163000838.html)

IDC puts a number on the AI skills gap: $5.5 trillion

Workera's breakdown of a new IDC report estimates the global AI skills gap will cost $5.5 trillion in 2026 alone, through delayed products, quality problems, lost revenue, and eroding competitiveness. More than 90% of organizations expect to face a critical skills shortfall. The detail that stands out is not the top-line dollar figure, it is what Workera's own assessment data shows underneath it: across more than 88,000 evaluations, only 13% of enterprise employees demonstrate the critical skills needed to work with AI agents, and 87% lack the judgment and task-proficiency to reliably verify or redirect what an AI system produces in production. Only a third of organizations consider themselves fully ready for AI-driven work.

That 13% figure is a sharper number than the headline $5.5 trillion, because it points at exactly what is missing: not access to AI tools, but the human judgment to check their output. It is a demand signal for performance-based validation at scale, not another course completion certificate. (Source: https://www.workera.ai/blog/the-5-5-trillion-skills-gap-what-idcs-new-report-reveals-about-ai-workforce-readiness)

A 20-minute computerized test predicts military aptitude as well as a 3-hour exam

A study published in JMIR Formative Research, led by the University of Minnesota with support from the Minneapolis VA Health Care System and the Minnesota Army National Guard, and funded by the Office of Naval Research and NIH, followed 267 service members through a 20-minute, self-administered computerized assessment called BrainHQ. The result: it strongly predicts scores on the AFQT and ASVAB, the Department of Defense's core tests for service fitness and job placement, by measuring real markers of brain performance, including speed, accuracy, planning, reasoning, and decision-making, in a fraction of the time and cost of the standard 1.5-to-3-hour in-person exam.

This is a direct methodological signal in PowerTechs' own layer of the market: scalable, computer-based psychometrics are starting to outperform long, manual testing batteries at measuring cognitive aptitude and resilience, in the exact military context where that measurement already matters most. It's the same underlying principle that trace-based skill assessment inside a simulation is built on: observed performance predicts readiness better than a fixed-length written test. (Source: https://www.globenewswire.com/news-release/2026/07/06/3322381/0/en/A-Big-Advance-Measuring-Military-Vocational-Aptitude-and-Cognitive-Resilience-comes-from-Brain-Games.html)

An AI engine drafts the assessment, a human still signs off

PsyMetrics launched an AI-driven engine called Test Architect that builds pre-hire assessments automatically: feed it a job description, and it maps the extracted requirements against cognitive, behavioral, and skills benchmarks drawn from a 30-year validated psychometric library, then assembles a role-specific battery from calibrated items. Critically, the company built it around a strict human-in-the-loop philosophy. A consultant reviews, edits, and authorizes every assessment before it launches, and the product is explicitly positioned as a "defensible" alternative to ungoverned AI recruiting tools.

This is a competitive methodology signal in the same layer PowerTechs operates in: assessment generation is automating fast, but the value that survives is the link to validated science plus a human checkpoint, not raw AI output alone. Trust in a skills assessment still runs through both machine speed and human accountability. (Source: https://www.globenewswire.com/news-release/2026/07/13/3326119/0/en/The-Antidote-to-Hiring-Bias-PsyMetrics-Launches-AI-Test-Architect-to-Anchor-Generative-Assessment-Building-in-30-Years-of-Psychometric-Science.html)

Takeaway

A business magazine says AI's $700 billion infrastructure boom is bottlenecked by skilled people, not chips. An analytics firm puts the resulting gap at $5.5 trillion, with only 13% of employees able to verify AI output. A military-funded study shows a 20-minute computerized test can predict aptitude as well as a three-hour exam. And a psychometrics vendor is proving that AI can draft an assessment fast as long as a human still signs off on it. Four unrelated sources, one shared direction: the market is racing toward faster, computer-based, evidence-backed ways to prove who is actually ready, and veterans with hands-on experience are sitting exactly where that demand is heading.

For veterans: see roles that match your military experience

For employers: book a 30-minute call