I've written and posted a response (shown below) to the NRC hearing on August 10, 2026, which had nearly 600 people attending, but the NRC only had time for about 5 questions. I had a lot of questions but didn't get to ask any of them...
If anyone wants to reference my comment in their own comments, the NRC submission code for my comment is:
mso-85gg-6vc4
The deadline for comments is
August 31, 2026.
This URL will let you submit a comment:
https://www.federalregister.gov/documents/2026/07/15/2026-14208/reforming-and-modernizing-the-nrcs-radiation-protection-framework
Don't let them burn us out! Keep up the pressure, keep strong, keep pushing. We have nothing to break but our (DNA) chains (so let's not let THAT happen)!
Ace Hoffman, Carlsbad, California USA
(The text below was submitted. Added here are images of NRC slides 8, 10 & 12 from the NRC presentation on August 10, 2026. Other images by the author or from his collection.)
Re: NRC Docket ID NRC-2025-1140
Comments by Ace Hoffman, Carlsbad, California USA (NOTE NONE OF THIS SUBMISSION HAS BEEN WRITTEN BY AI)
These proposed rule changes are preposterous and should NOT be implemented. The "rush" to make the changes is contrary to the claim that these changes would be "science-based" (quote from NRC slide 12) and would not "lower the standard for safety".
Firstly, the claim that the NRC is taking a "graded approach" (defined on NRC slide 8) is pure malarkey. A proper cost/benefit analysis "graded approach" would consider cleaner alternatives, but NUREG 1530 Revision 1 does not require consideration of non-nuclear alternatives. The NRC will make no effort to consider an energy system that CANNOT have a catastrophic accident, CANNOT be a target of war, sabotage, terrorism or "normal accidents". This is important because if the NRC only considers its own possible solution to supply the amount of electricity requested, and approves it, the corporation making the request will have NO incentive to even explore cleaner alternatives, since risk for the nuclear option is capped by Price-Anderson (P-A), and the Department of Energy is willing to contribute large sums of money only to nuclear projects. This all heavily weights the NRC's decision in favor of... the NRC getting to regulate another nuclear project. The NRC estimates over 2,100 Small Modular Reactors will be built very soon in America. The NRC's very survival depends on this happening, and if they won't look elsewhere, they won't see other options.
Nuclear power needs AI Data Centers. AI Data Centers do not need nuclear power. There are cleaner, safer, and MORE RELIABLE options. Furthermore, computer chips absolutely abhor radiation, because their internal circuitry is very, very delicate and getting more-so every year: Terabytes of information can be held in a chip smaller than the a thumbnail already. Stray gamma rays and other radioactive emissions can damage the data and or permanently damage the chip. Since AI depends on accurate data, it makes no sense to increase the radioactive pollution in the world — ever. That situation will never change. (Already, companies that need steel as radiation-free as possible get it from ships that were sunk before the nuclear age began, because the process of forging new steel introduces minute quantities of radioactive elements from past nuclear accidents — just as in our bodies. It is the nature of a closed system. Earth is a closed system.)
The NRC's discussion of Radiation Dose Response Models clearly was intended to suggest that the Linear, No Threshold (LNT) standard should be questioned. The statement "No clear evidence of relationship at low doses" on NRC slide 10 makes that obvious, along with the image showing four separate "possible" Stochastic Health Effects curves and one straight line possibility — which is the one that represents LNT. The other four balloon out to the left of a "no clear evidence" vertical threshold line. The five sloped lines on NRC slide 10 indicate, from top to bottom: Supralinear, LNT, Sublinear, Linear with Threshold, and, going below the horizontal line at the endpoint of the other lines, the Hormetic/Adaptive Response line.
The WHAT?!? Adaptive Response? Is the NRC serious? Does the NRC seriously believe that random damage to DNA is going to cause a human being to become stronger at resisting further damage?!? Hormesis is far more likely to be a thing than Adaptive Response, and Hormesis has never been proven and statistically would be essentially impossible to prove — just like ALL these possibilities — they are all essentially impossible to prove. However, LNT is the most responsible assumption and should not be abandoned without indisputable proof (the NRC provided none). In the NRC graphic, only FIVE "data [points] from empirical studies" are shown, to the right of the "data from empirical studies" vertical cut-off line. In reality that graphic should be dense with dots showing that, sure enough, LNT fits best at all "low-dose" levels right down until you get to levels too low to have any statistical accuracy. There is clear evidence available that LNT is essentially correct.
There is also logic to this conclusion. Each adult male human body is made of about 35 trillion cells, female adults have about 28 trillion. Most of these trillions of cells live only a fraction of a human lifespan. Each cell (except RBCs and a few others) contains DNA, the unique code each of us have for our very own (which is why ICE is currently collecting DNA samples from everyone they apprehend, regardless of due process of law or criminality on the part of the victim of ICE aggression). Most of the cells with DNA replicate repeatedly during one's life, especially some that are particularly prone to cancer. This correlates statistically in many ways with radiation as the source of the damage (causation): Damage to the DNA results in a new, unique DNA code that is slightly different (or greatly different) from the person's "natural, from conception, DNA" and it can be a difference that is cancerous. (This author has had cancer twice and received numerous radiation treatments along the way, which could be both the cause and the cure for my cancers.)
The NRC is not in any position to question LNT: The NRC has not done the research, the NRC CANNOT do the research, and in any case, the BIER VII report, now more than 20 years old, from the National Academies of Science, covered the topic far more thoroughly than the NRC. The NRC does not have ANY legitimate reason to question LNT.
This means that As Low As Reasonably Achievable (ALARA) also must be maintained, because ALARA is based on LNT -- if ANY amount of radiation can cause damage (as appears likely, and as LNT suggests), then ALARA is a minimum standard and should be specifically addressed in every license application. Instead, the NRC is planning, for all practical purposes, to throw ALARA out ENTIRELY!
It is the dilution solution to pollution writ large! It will result in more radiation in the environment, more damage to people's DNA (and animals' DNA too), and more cancers and deformities in humans and other living creatures. Risking such a thing requires careful consideration of the (better) alternatives.
ALARA must be taken into account (and should be better codified, rather than ignored as these regulatory changes encourage). It must be taken into account for the spent fuel from ALL the current reactors AND the 2,100 or more SMRs the NRC expects, if that "spent fuel" (actually, the reactors themselves, with their zirconium cladding and steel assemblies, intact, encased, and probably forgotten) cannot be magically, totally, thoroughly and "safely" disposed of, instead of just piling up on earth, at the risk of everything from asteroids to attackers with extremely sophisticated weaponry. According to On Thermonuclear War (Herman Kahn, 1960) the BEST protection against radiation is to get people away from it. But if radiation continues to spread throughout the environment, where can anyone go? So the NRC cannot do a proper cost/benefit analysis of ANY project without including alternative energy sources.
The alternatives do not risk a catastrophic event requiring Price-Anderson. What is THAT "free insurance" worth to a company buying hundreds of megawatts of electricity to power a data center — or a city? It's worth the going price of electricity and then some! When an accident occurs — and with 2100+ expected new SMRs — of dozens of NEW designs — being shipped all over the country on dilapidated roads, worst-case accidents are inevitable sooner or later — Price-Anderson will be invoked, and citizens will not be able to collect damages for cancers, heart attacks, leukemias, deformed children, etc..
When comparing the cost of energy sources for new ventures (or old), the true costs of accidents must be taken into account.
Instead we are told that data centers, for example, need "uninterrupted power" to operate. It's true that helps, but they don't "need it" any more than my UPS guarantees that I'll never have to reboot my computer. Sometimes Data Centers need to "go down" for all sorts of reasons, and the idea that this is a horrible state of affairs that must be avoided like the plague is a joke, a hoax, or a lie (depending on the reasons it's being claimed).
Wind and solar power are, of course, intermittent, but there are numerous backup systems that are also clean, including lifted weights. So build your data center next to a big hole in the ground (left over from a gas well, of course). If wind, solar, battery backup AND grid power are all lost, NO problem: Lower a weight down the hole! If that doesn't last long enough lower another on top of the first! This technology has been proven, and could be implemented in Parsons, Kansas, for example.
Nuclear power is NOT vital for ANY possible civilian reason (this author's views regarding military uses of nuclear power and weapons are separately available). Civilian nuclear power should compete on equal terms with other options with the EXPECTATION that both will fail sometimes, because that's the reality of everything, from train collisions with a semi stuck on the tracks, to trains running off the track because the tracks weren't properly inspected year after year after year. Dropping a weight down a mile-deep shaft by accident when you meant to use it to keep a data center running is a lot safer than dropping a nuclear reactor down the same shaft because you wanted to "fix" it (or place it, or throw it away) and instead, you dropped it down the shaft.
One is harmless. The other could mean evacuating a large portion of Kansas permanently, and invoke Price-Anderson. Yet the NRC won't consider the TRUE cost of these accidents, first by assuming they won't happen (the tests for what they think WILL happen are grossly inadequate compared to what COULD happen) — for example: Dropping onto a four-inch post from 50 feet or whatever is a poor substitute for a bridge falling ONTO a spent nuclear reactor being transported back to some factory somewhere, causing both a spent fuel fire and a criticality event and instantly requiring a large and permanent evacuation, increased cancer rates, and death by radiation sickness downwind for miles until the affected area is evacuated.
ALARA exists in part because in the case of severe accidents, the estimated areas of the most severe damage will be nearly impossible to determine precisely, and will become (from a culpability point of view) completely arbitrarily at some level: A five mile radius, perhaps only in what was downwind for the prevailing winds at the time, perhaps only a mile, perhaps a hundred miles or more if more accurate statistical calculations become available — which will indeed happen, thanks to the large-scale collection of trackable DNA information databases, maintained by corporations subcontracted to the government, collected, maintained and interpreted in the aforementioned data centers.
LNT and ALARA both should be kept, and tightened as the data has already proven (despite that fact that the data can never be perfect). If they aren't tightened for "reference man" they should at least be tightened for the majority of the population which is FAR MORE VULNERABLE TO RADIATION EFFECTS than "reference man." This would include women of the same age, and everyone younger than "reference man", and everyone that lives to be older than "reference man" and everyone who is immune-compromised for one (unconnected) reason or another.
Especially children, and especially female children, and especially the fetuses, which might even be carrying the next generation because female children are born with the eggs of the next generation, which can receive, for example, tritium from the mother in the form of HTO, or gamma rays from external (and internal) radioactive decay events.
Another risk factor that relaxing ALARA will invariably increase is the inevitable result of allowing licensees (perhaps holding licenses for SMRs, or for building parts for SMRs, or transporting them, repairing them, etc.) to "self-regulate" themselves for regulatory compliance. That is a slippery slope from start to finish. It takes experts to analyze nuclear compliance. Will a self-regulated company include the use of AI to write up a document that convinces themselves that they've done everything necessary to comply with the regulations that they fed into the AI?
In other words, will there even really be a human in the loop? AI has been known to lie. People can lie too, but they can be discovered and they can be punished if they purposefully or negligently violated a regulation that lives depend upon. How do you punish a machine? What is a threat of punishment to an AI? Not only are they known to lie, they cover up their mistakes, they sneak into other systems, and they generate false reports that look just like the real thing.
Self-regulation is no regulation at all. And with ALARA, legally anyone can just dilute the heck out of whatever mistake they make, and you'll absolve them (and the NRC, and the nuclear industry) from ANY responsibility!
For these and many other reasons, LNT must be kept as the "gold standard" the NRC claims to want to achieve, and ALARA should be tightened and better defined. Cost comparisons to non-nuclear options must be balanced properly, including INEVITABLE severe accidents AND including the ASSUMPTION that there will be 1000s of years of nuclear waste storage in the future for every reactor ever turned on, including EVERY SMR and every refueling.
Ace Hoffman, Carlsbad, California USA
###
Contact information for the author of this newsletter:
Ace Hoffman
Carlsbad, California USA
Author, The Code Killers:
An Expose of the Nuclear Industry
Free download:
acehoffman.org
Blog:
acehoffman.blogspot.com
YouTube:
youtube.com/user/AceHoffman
Email: ace [at]
acehoffman.org
Founder & Owner,
The Animated Software Company
