Volts
Volts

Best of August 2026

This is a free preview of a paid episode. To hear more, visit www.volts.wtf Volts published eight conversations in August — with two Idaho state legislators, a grid cybersecurity expert, the architect of Uruguay's energy transition, an EV charging exec, and more — which is a lot for anyone to l

Topics Discussed

Episode Summary

Executive Summary: This Volts best-of episode introduces a new monthly roundup format and highlights two major themes from August: how data center growth shifts grid costs and risk onto ratepayers across different market designs, and how concerns about Chinese-made inverters should be understood in practical, non-hyped terms. The episode emphasizes who bears risk, who pays, and what controls are realistically available.

Main Topics: Monthly best-of format and listener-supported model (Priority: 3/5): David Roberts explains that the show will now release a monthly compilation of August episodes, with paid subscribers getting early access and older editions eventually becoming free. He frames it as a listener-requested way to help audiences catch up on missed content. Data centers and cost/risk allocation across market structures (Priority: 5/5): Salim Chapman of Climate Cabinet argues that data center-driven grid costs create similar problems in vertically integrated, restructured, and ERCOT-style markets: ratepayers absorb the risk when demand forecasts are wrong or when grid upgrades are triggered by new load. Georgia as a vertically integrated utility example (Priority: 4/5): In Georgia, a major gas expansion was approved using developer supply forecasts. The decision followed normal process, but if demand falls short, ratepayers can be stuck with stranded asset costs. PJM capacity market spillovers (Priority: 4/5): In PJM, a single hyperscale campus can be large enough to trigger a capacity market event, raising costs across the 13-state footprint even though many customers had no role in approving the facility. ERCOT volatility and extreme weather (Priority: 4/5): In ERCOT, hyperscale growth interacts with extreme weather and emergency dispatch measures, creating billions in surplus charges and market price volatility that do not appear as a clear line item tied to data center growth. Chinese inverter security: separating hardware from software risk (Priority: 5/5): Patrick Miller explains that many suspicious-looking components in Chinese-made inverters may simply be standard commodity board features, while the more serious concern is software-level control, update integrity, and supply-chain oversight. Practical limits and controls for inverter security (Priority: 4/5): Miller argues that thousands or millions of installed devices cannot simply be removed, but software can be governed through code inspection, update review, and transparency requirements, reducing the risk of remote abuse.

Key Arguments: Data center load growth creates a recurring pattern: utilities and market operators respond to developer forecasts, but ratepayers often bear the cost if those forecasts prove wrong. This problem is not limited to one electricity market design; regulated utilities, capacity markets, and ERCOT can all produce similar cost-shifting outcomes. In Georgia, the process worked as designed, but the structure still exposed ratepayers to stranded gas infrastructure risk if load does not materialize. In PJM, large new loads can ripple across a multi-state market, meaning people far from the data center may pay for local capacity needs they did not approve. In ERCOT, the combination of hyperscale load and extreme weather can generate system-wide volatility and emergency costs that are hard for individual customers to trace. Concerns about Chinese inverters should distinguish between benign commodity hardware components and genuinely concerning software/communications pathways. Existing installed inverters cannot realistically be audited or removed at scale, so the practical solution is to control software updates, code transparency, and device management processes. A component being present on a board does not mean it is automatically active; it generally must be configured and provisioned before it can function. The core security challenge is not just whether a chip exists, but whether there is a controlled path for activation, communication, or unauthorized updates.

Data Points: Episode date: September 4, 2026 - David Roberts introduces the best-of August compilation episode. Monthly compilation format: 1 new best-of episode per month - The show will now regularly publish a monthly roundup with early access for paid subscribers. Georgia market type: Vertically integrated / regulated - Used as the example of a traditional utility structure in the data center discussion. PJM footprint: 13 states - Chapman says PJM capacity costs can spread across the full multi-state footprint. Local peak demand share: 30% - A single hyperscale campus in the PJM example reached about 30% of local peak demand. Installed inverter count: Millions - Miller says there are millions of Chinese-made inverters and related devices already in service. Product sourcing scale: Commodity boards bought in millions - He describes manufacturers buying generic boards in very large quantities. ERCOT cost impact: Billions in surplus charges - Miller/Roberts discuss emergency programs and gas dispatch tied to hyperscale growth and extreme weather.

Pivotal Quotes: "who is taking on the risk and who's capturing the value" — David Roberts: Summarizing the core equity issue in the data center and grid-cost discussion. "the process worked exactly as it should" — Salim Chapman: Describing how the utility and market mechanisms functioned normally even while producing problematic cost exposure. "you can't yank them all out" — David Roberts: Framing the practical challenge of dealing with already-installed Chinese inverters at scale.

Implications: Utilities, regulators, and grid operators need better rules for assigning data-center costs and stronger software-focused security controls for inverters. The episode suggests the main challenge is not technology alone, but governance: who pays, who decides, and who can be trusted to update devices.

🔓 Sign Up for Unlimited Episode Search

About Volts

View all episodes from Volts