Infrastructure, sovereignty, governance and security—not model performance alone—will define the next phase of AI.
Eight months into 2026, the AI landscape is evolving broadly as we anticipated—but the strategic question has shifted. The issue is no longer simply who has the most capable models. Competitive advantage increasingly depends on who can assemble, finance, secure and govern the full AI stack—from compute and energy to models, infrastructure and regulatory access—and deploy it reliably at scale.
In this mid-year assessment, DGA revisits ten predictions made at the start of 2026 and examines the forces likely to shape the market through 2027. The report assesses how AI agents are deepening demand for inference; U.S. infrastructure expansion is encountering physical and political constraints; sovereign AI initiatives are spreading even as frontier capability remains concentrated; and competing regulatory, open-model and national technology ecosystems are creating new operational and compliance risks.
For business leaders, the implication is clear: AI strategy can no longer be separated from infrastructure, jurisdiction, security, vendor dependence and public policy. The full report identifies the developments to watch—and the strategic questions organizations should be testing now.