This week shows three versions of the same race. China is subsidizing electricity to keep its AI ambitions alive despite chip bans. Stateside, utilities admit the digital grid won’t be secured by software alone—it needs people who know OT and cyber. And in the middle of the map, SPP proposes a structural fix: merge transmission planning with interconnection so steel and silicon arrive together. Different playbooks, same objective—more reliable power, faster.

China’s cheap watts for AI

What happened: Local governments in China are juicing electricity subsidies—reportedly halving bills for some hyperscale data centers—to help national tech champions (ByteDance, Alibaba, Tencent) offset higher costs after curbs on Nvidia AI chips.

Why it matters: This is industrial policy by the kilowatt. If your AI stack can’t import the best silicon, you lower the cost of everything around it—power, land, permits—to keep training runs economical. It’s also a reminder for U.S. regulators: price signals and connection speed beat slogans. If we want domestic compute to root here, we need transparent interconnection terms, flexible-load contracts, and firm, dispatchable supply—not a patchwork of bespoke favors. Free-market takeaway: compete on rules that scale (siting, tariffs, contracting), not subsidies that don’t.

Black & Veatch: people > point solutions on grid cyber

What happened: In its 2025 Electric Report, Black & Veatch says utilities are prioritizing cybersecurity training (≈40%) ahead of buying more tools. Only ~22% run unified OT/IT teams; many still don’t see where physical and cyber risks intersect, even as substation attacks rise. The core gap isn’t tech—it’s orchestration across ADMS, DERMS, comms, and OT security.

Why it matters: The digital grid fails at the seams. Buying another dashboard won’t fix misaligned org charts or a thin bench of OT-fluent defenders. The winning utilities will formalize consequence-driven risk (protect what can hurt people and keep lights on), unify OT/IT authority, and make incident response as routine as storm prep. That’s slower to brag about than a new platform—but much harder to hack.

SPP’s big swing: merge planning with interconnection

What happened: Southwest Power Pool filed an “unprecedented” Consolidated Planning Process at FERC, fusing long-term transmission planning with the generator interconnection queue. A new GRID-C fee would give interconnecting resources clearer, upfront cost shares while spreading regional upgrades more predictably.

Why it matters: Queue reform without network build is theater; network build without cost clarity is politics. SPP’s move tries to solve both—plan the backbone, price it transparently, and stop endless restudies. If approved, expect faster time-to-wire for thermal, storage, and renewables alike—and a template other RTOs will be pressured to copy. Markets love certainty.

Upgrade to Grid Brief Premium to get extra deep dives into energy issues all over the world.

Conversation Starters

• The Economist: Solar in China has become too big to fail.
  China’s industrial solar machine now rivals its state banks in strategic weight. With automation, scale, and near-total control of polysilicon and wafer supply, Beijing’s dominance is structural, not cyclical.
  Why read: This is the real “China shock 2.0”—not in consumer goods, but in clean tech. Any U.S. reshoring strategy that ignores the manufacturing moat built around sunlight is wishful thinking.

• CNBC: Trump’s nuclear power investment in Westinghouse could lead to an IPO—with the U.S. government as shareholder.
  A new $80 billion plan between the administration, Cameco, and Brookfield could spin Westinghouse into a publicly traded company—partially owned by taxpayers.
  Why read: Washington’s first direct equity stake in a reactor builder since the Cold War would make nuclear not just a policy priority, but a nationalized asset class.

• Roll Call: Worker shortage looms over new U.S. nuclear power focus.
  DOE wants to quadruple nuclear capacity in 25 years—but the workforce would need to triple first. With retirements rising and apprenticeships lagging, the “renaissance” faces a manpower recession.
  Why read: No number of small modular reactors will matter if there’s no one left to weld, wire, or run them. The bottleneck isn’t the atom—it’s the human.

Good Bet: Utilities that lock in flex-load deals with data centers (curtailment rights + advance payments for upgrades), while standing up unified OT/IT cyber teams. That combo lowers customer risk, speeds connections, and hardens the digital perimeter.

Bad Bet: Assuming you can “tool” your way to security—or that private, off-grid plants make hyperscalers grid-independent. In practice, redundancy, market access, and NERC obligations pull you back to the network. Ignore org design and interconnection reality at your peril.

Share Grid Brief

We rely on word of mouth to grow. If you're enjoying this, don't forget to forward Grid Brief to your friends and ask them to subscribe!