“US rate payers wasting $12 billion of money that they don’t have to waste because PJM is making a modeling mistake”
Clip via Jordan Nanos with Robert Boswall (@RobertBoswall @JordanNanos) / SemiAnalysis — Something Else Weekly — Ep. 026 – PJM’s $12B Modeling Mistake Is Hitting Ratepayers Again (Datacenter, Energy).
The AI buildout has a grid problem. And the grid problem has a math problem.
On a recent episode of SemiAnalysis’s Something Else Weekly, Jordan Nanos and Robert Boswall surfaced an argument that deserves far more attention than it’s getting: America’s largest grid operator may be burning ratepayer money at scale — not from corruption, not from underinvestment, but from a modeling error.
That’s a specific and consequential claim. Here’s the structural read on why it matters.
The Argument
Nanos and Boswall’s position — as expressed in this episode — is that PJM’s modeling approach is generating a systematic error. The result, in their framing: $12 billion in unnecessary costs landing on US ratepayers who have no visibility into why their bills reflect that burden.
“Money that they don’t have to waste” — the phrasing is precise. This isn’t an argument about necessary costs of transition. It’s an argument about avoidable costs from an analytical failure.
— FourWeekMBA analytical read on the SemiAnalysis clip
Structural Read — FourWeekMBA Analysis
In our Permission Layer framework, the grid operator functions as an invisible regulatory chokepoint — the entity whose models determine what gets built, what gets priced, and who pays. When that model is wrong, the error doesn’t surface as a line-item. It surfaces as higher bills, stranded investment, and misallocated capacity.
The datacenter and AI infrastructure boom makes this load-forecasting problem sharper, not softer. More demand volatility means modeling errors compound faster — and at larger scale.
The Numbers In Play (Per Episode Claim)
$12B
Alleged unnecessary cost to US ratepayers (per Nanos/Boswall)
PJM
The grid operator at the center of the modeling argument
Why This Surfaces Now — FourWeekMBA Analysis
The AI infrastructure conversation has focused almost entirely on chips, models, and data. Relatively little analytical attention has gone to the grid layer that powers all of it. If the argument Nanos and Boswall are making holds up to scrutiny, the story isn’t just about one operator’s methodology — it’s about whether the institutions governing electricity markets are calibrated for a world of massive, irregular datacenter load. That’s a structural question, not a cyclical one.
Grid modeling is not a boring back-office problem. If the SemiAnalysis argument is right, it’s a $12 billion tax on US households — imposed not by policy, but by analytical error. The AI buildout will stress these systems further. The question worth asking: who is actually auditing the models that set the price of electricity?
Clip via Jordan Nanos with Robert Boswall (@RobertBoswall @JordanNanos) / SemiAnalysis — Something Else Weekly — Ep. 026 – PJM’s $12B Modeling Mistake Is Hitting Ratepayers Again (Datacenter, Energy).
This is editorial and structural analysis based on a podcast clip. The $12B figure and all claims about PJM’s modeling are the speakers’ argument as expressed in that episode — not independently verified facts, and not investment advice.
Gennaro is the creator of FourWeekMBA, which reached about four million business people, comprising C-level executives, investors, analysts, product managers, and aspiring digital entrepreneurs in 2022 alone | He is also Director of Sales for a high-tech scaleup in the AI Industry | In 2012, Gennaro earned an International MBA with emphasis on Corporate Finance and Business Strategy.
Scroll to Top
Discover more from FourWeekMBA
Subscribe now to keep reading and get access to the full archive.
