Another postcard arrived this week, and it’s a considerable step up from the last one. Here it is in full, both sides — everything below is quoted from it, and you can check me.
Glossy, two-sided, every household. One full side is given over to cancer surgery — two photographs of an operating room, and a story about U-M researchers developing a tool that analyzes tumor tissue in ten seconds, cutting the rate of patients with missed cancer tissue from 25% to 4%.
Underneath, in italics: U-M’s proposed research computing center would expand the advanced computing resources that help make research like this possible.
Then, across the bottom: Flip over to see how U-M is powering the next life-saving breakthrough.
I want to be fair about the research. That cancer work is real, it’s impressive, and it deserves the attention. I have no quarrel with it at all.
I have a quarrel with what the card does next.
Count the words
There are roughly 265 words of copy on this postcard, across both sides.
Four of them are “Los Alamos National Laboratory.” The phrase appears exactly once, at the end of a sentence, in a subordinate clause:
Expanded computing capacity is needed to enable future breakthroughs. That is why U-M is planning to build a research computing center in Ypsilanti Township with Los Alamos National Laboratory.
That’s it. That is the entire disclosure, to every household in this township, of the fact that about 91% of this facility’s computing capacity belongs to a federal nuclear weapons laboratory.1
The word “cancer” appears five times.
The words nuclear, weapons, classified, megawatt, gallon, substation and generator do not appear at all.
“An academic research center”
The card has a box headed Quick Facts About the Project. Five bullets. Here they are in full:
- An academic research center
- Focused on scientific innovation
- No impact on local utility rates
- Safe and environmentally conscious
- Protecting local water resources
Start with the first one, because it’s the one that matters most.
The University’s own filing to the State of Michigan describes 100 megawatts of capacity for Los Alamos and a separate 10 megawatt academic computing center for the University’s non-classified research.1 The project leader confirmed roughly that split aloud at the September 9 town hall.2 The University’s own FAQ describes a 240,000 square foot federal building including classified research, next to a 50,000 square foot academic one.3
Nine percent of the capacity is academic. The card calls the whole thing an academic research center
and lists it as fact number one.
I don’t know how to read that as anything other than deliberate.
“No impact on local utility rates”
This one is presented as a quick fact. It is a prediction, and it isn’t the University’s to make.
Whether a large new load shifts costs onto other ratepayers is decided by DTE’s rate filings and the Michigan Public Service Commission. Not by the University, and not by a postcard.
And the University’s own Ford School published a brief in July 2025 — jointly with the Michigan Environmental Justice Coalition — describing how utilities routinely pass infrastructure and procurement costs to residential customers as demand climbs.4 A statewide survey from the same school, published last month, found 57% of Michigan local officials expect data centers to harm utility bills in their communities.5
If the University is confident enough to print this on a postcard, it should be confident enough to commit to it in a form that binds — a rate case commitment, or language in an agreement with this township. A bullet point on a mailer costs nothing and guarantees nothing.
“Protecting local water resources”
Two years. Three public meetings. A water authority that adopted a moratorium to protect its own capacity.
Still no figure in gallons per day. Not an estimate, not a range, not a cap.
I’ve written at length about why that number depends almost entirely on one engineering decision — how the facility rejects its heat. Dry cooling uses essentially none. Evaporative cooling can consume over a million gallons a day at this scale. The University has said the cooling design isn’t settled.
So protecting local water resources
is a sentiment printed where a number should be.
“Safe and environmentally conscious”
This is the bullet I’d most like someone at the University to defend.
The University has filed two Baseline Environmental Assessments with the State recording all three Textile Road parcels as “facilities” under Part 201 — meaning contamination above generic residential cleanup criteria. Mercury above the volatilization-to-indoor-air criterion on the twenty-acre parcel. Lead at 24,000 µg/kg. PCBs. Arsenic and chromium across the larger parcels.6
Filing those assessments was the correct thing to do, and I’ve said so. The University did its environmental diligence properly, before each purchase, both times.
It has never mentioned any of it to this community. Not at a town hall, not in the FAQ, and not on a postcard that describes the project as environmentally conscious.
I found out by asking the State.
“Your Voice Matters”
In a blue box at the bottom right: U-M is committed to community input. Share your priorities.
I’ll let the record speak.
The Township supervisor learned the land had been bought by text message, after the closing. The Township’s formal opposition, its resolutions, its request through a state senator and its appearance before the Regents have produced no change in the plan. At the first town hall, the panel confirmed it had no authority to change or stop the project. At the second, the supervisor had to be chanted back onto the stage by the room in order to speak at all.
Community input has been offered continuously for two years. What’s missing isn’t input. It’s any evidence that it changes anything.
The thing I keep coming back to
If this project is as good for Ypsilanti Township as the postcard says, it should survive being described accurately.
A version of this card that led with a 110-megawatt computing facility, about 91% of it for a federal national security laboratory, including classified research
would be truthful. It would also be a harder sell, which is presumably why we got cancer surgery instead.
And there’s a real question underneath the cancer framing that nobody has answered. The tumor-analysis tool described on the card already exists — it’s being used in operating rooms now. The card says the new facility would expand the resources that help make research like this possible, which is carefully worded and probably true of the ten academic megawatts.
It tells you nothing about the other hundred.
What I’d ask for
The card gives two addresses — [email protected] to request a meeting, and [email protected] to share priorities. Both are real channels and I’d encourage people to use them.
If you write, here is what I’d ask:
- How is a facility with 100 of its 110 megawatts allocated to Los Alamos
an academic research center
? - On what basis does the University state there will be no impact on local utility rates, and will it commit to that in a binding form?
- What is the projected consumptive water use in gallons per day?
- Why has the community never been told that all three parcels are recorded with the State as contaminated facilities under Part 201?
Ask one. Ask for a number or a date. The same rules apply in writing as at a microphone.
Sources
- MEDC memorandum to the Michigan Strategic Fund Board, December 10, 2024. Board packet (PDF)
- Public recording of the September 9, 2026 town hall. youtube.com · my account, with timestamps
- University of Michigan Office of the Vice President for Research, LANL FAQ. research.umich.edu
- Nguyen and Green, What Happens When Data Centers Come to Town?, Ford School of Public Policy, with the Michigan Environmental Justice Coalition, July 2025. PDF
- Horner and Leiser, Michigan local government leaders’ views on large-scale data center development, Michigan Public Policy Survey, CLOSUP, Ford School of Public Policy, September 2026. Full report (PDF) · my summary
- Baseline Environmental Assessments filed with EGLE, January 2025 and March 2026. My analysis, with both filings linked in full
The postcard is published here in full (PDF). Word counts are my own, taken from the complete text of both sides of the mailer received October 2026. Quotations are transcribed from the card itself. If I’ve miscounted or misquoted, tell me and I’ll fix it and say that I did.
