The Digital Bridge with Lawrence Eta

From Chiba to Ogun State, AI Data Centers Are Leaving the Grid. Norway Shows Why.

Japan's JERA plans a 400-megawatt AI data center beside its own power station. A Nigerian energy group is pairing a 20-megawatt facility with a 100-megawatt gas plant built for it. Australia's Project Ares would run off-grid on solar, gas, and 16 gigawatt-hours of batteries. In Norway, where data centers with approved connections could use 25 terawatt-hours by 2035 against a typical 15 terawatt-hour surplus, parliament has just debated halting them. Whoever supplies the power now sets the terms.

The Fine Print Decides Who Gets AI Capacity: Brazil, the EU, and India

The AI Stack: From Ownership to Institutional Control

Some Countries Are Already Running Sovereign AI Infrastructure. Others Are Still Writing the Rules

AI Sovereignty Is Not Independence

The real strategic question is whether countries and organizations can retain choice when access to technologies they do not control begins to change.

There Will Not Be One Global Model for AI. There Are Already Five

South Korea is running a government tournament to build its own frontier models. India has put 38,000 subsidized GPUs into the hands of startups and researchers instead of building a national champion.

Two-thirds of CIOs and CTOs are accountable for AI systems they do not fully control.

A global IBM study of 2,000 CIOs and CTOs found that two thirds carry accountability for AI systems they cannot fully govern. A separate survey of 900 CEOs found 87 percent would stake their job on AI delivering results, while 34 percent still will not let it make a decision without a human signing off first. Those two numbers, read together, are the real story of leadership in September 2026.

The Geography of AI Power Is Changing

Saudi Arabia, the UK, France, India and the EU are investing deliberately in AI infrastructure, compute and domestic capability. The objective is not to replicate Silicon Valley. It is to determine where strategic capability must reside and where dependency creates risk.

McKinsey Says AI Is Finally Paying Off. Only 6 Percent Are High Performers.

Companies Bet AI Would Reduce the Need for People. The Evidence Is Forcing a Rethink

The AI Layoff Wave Is Already Reversing. Gartner Says Half of It Gets Rehired by 2027.

Boards Just Got a Gap Report on AI Oversight. It Names the Same Condition Every Failed Transformation Skips.

From Aspiration to Delivery: Creating the Conditions for Transformation

The AI Decision That Boards Can No Longer Delegate

The next phase of AI adoption is not primarily a technology challenge. It is an accountability challenge.