Review of January, 2026 Bulletin of the Atomic Scientists' article by Adam Stein and PJ Seel titled: "No, the United States does not need a costly national cancer study near nuclear reactors".
Link to Stein & Seel article:
https://thebulletin.org/2026/01/no-the-united-states-does-not-need-a-costly-national-cancer-study-near-nuclear-reactors/amp/
Background:
Stein and Seel's article attempted to rebut a September, 2025 Bulletin of Atomic Scientists' article by Joseph Mangano and Robert Alvarez:
https://thebulletin.org/2025/09/why-a-national-cancer-study-near-us-reactors-must-be-conducted-before-any-new-expansion-of-nuclear-power/
January 4, 2026
Review by Sharon and Ace Hoffman
Stein and Seel's January 2, 2026 Bulletin of the Atomic Scientists' article seems to have been written just to create confusion regarding the dangers of radioactivity in the environment. It brings nothing new to the table.
For example, the authors make the following perfectly reasonable statement: "[R]adiation exposure doses from nuclear power plants are further confounded by risk factors such as smoking, radon, air pollution, pesticides, diet, and workplace exposures, each individually contributing to a far larger extent to cancer incidence than routine reactor emissions possibly would."
In reality, the risk from each of the "confounding factors" they mentioned can be enhanced by excess radiation in the environment. Thus, radiation confounds the confounding factors! And yet, Stein and Seel conclude that further research concerning radiation from routine reactor emissions has no value.
Using the same argument about "confounding factors," Stein and Seel reject results from INWORKS and other studies that found elevated rates of cancer and other diseases in nuclear workers, claiming such studies failed to adjust for lifestyle factors. (See Ian Fairlie's excellent review of the INWORKS results in the links below.)
Stein and Seel assume that the United States needs more reactors and less regulation of radiation in the environment -- but the fact that the authors have significant business interests in new reactor projects is not disclosed in the article. (To find this glaring omission, one has to click on the authors' names, and then look up the company both authors work for.)
Stein and Seel don't completely reject the Linear No Threshold (LNT) theory, stating: "[T]here is no other model available that can provide more clarity to explain the highly variable data in the very low dose region". However, they reject the use of LNT to regulate radiation exposure, suggesting that because there are so many other causes of cancer, there is no reason to study potential health impacts from operating nuclear power plants. None of this makes any sense, if for no other reason than the fact that many of us will get cancer and will need enormous amounts of radiation exposures to "cure" or remove it -- and we'll probably all get numerous x-rays during our lifetime. Stein and Seel treat LNT as if it only relates to extremely low- vs no- doses of radiation, rather than additional, cumulative doses on top of levels that are already well-known to be harmful (even when they are considered medically necessary). Increased radiation levels in the environment do no one any good, even ignoring any "confounding factors."
Most importantly, the LNT "model" ALSO fits what science has learned, since the dawn of the nuclear age, about DNA and its physical and chemical structure, about cancer and its "starter cells" which contain damaged DNA, and about a thousand other things regarding how radiation can damage living organisms.
Stein and Seel fail to address numerous issues about low-dose radiation, such as the differences between internal and external radiation effects. Instead the authors question the accuracy of previous studies of radiation impacts on health, and link to a study that uses Causal Machine Learning (aka Artificial Intelligence) to analyze existing data on low-level radiation exposure. Their premise seems to be that we have more than enough data, despite the fact that they point out many weaknesses in that data.
The main reason to study the local effects of nuclear energy production is to estimate the global effects on humans and every other life-form on earth, forevermore, from everything that can create or spread radiation in any way.
Stein and Seel take it as a given that there should be MORE nuclear power plants, and thus, it is a given that in their view, billions of people need to accept additional, randomized radiation doses on top of everything they are bound or likely to get anyway. Furthermore, everyone, even far into the future, should accept anyone else's additional radiation without receiving any benefit from it themselves, just because that other person, perhaps eons ago, got a momentary benefit from the electricity while producing the radioactive waste or effluent.
Stein and Seel envision a multi-trillion dollar industry, but don't want less than a thousandth of that amount invested to see what level of safety is actually required. There is already ample proof the nuclear industry is a failure on many other fronts, so to that extent we agree there is no need for another study, but even if the industry were to shut down today, there would still need to be better studies of low-level radiation effects.
There are numerous radioisotopes with long half-lives that are released from "properly" operating reactors. Over time, these "hot" particles travel well beyond any local area. The harm they do in the environment is orders-of-magnitude more difficult to measure than any local effects, but LNT assumes that ANY effect found in a local study should be proportionately applied to the rest of the global environment. Therefore, ALL accidents and releases must be considered: TMI, Chornobyl, Fukushima, Santa Susanna, Mayak, Sellafield, etc., and the probability of future accidents, both to operating reactors and to spent fuel and the entire nuclear fuel chain. If a little additional radiation truly isn't worth worrying about (that is, if Stein and Seel are correct), that alone doesn't validate the continuation of the nuclear industry -- not by a long shot.
Stein and Seel say that according to a 1996 study: "[T]here was a chance for 4,000 radiation-induced cancer deaths [from Chornobyl]". They also quote a 2006 statement from UN organizations including WHO which says: "It is impossible to assess reliably, with any precision, numbers of fatal cancers caused by radiation exposure due to Chernobyl accident."
Once again Stein and Seel say something that is true and use it to infer an unsupported conclusion. In this case, while the WHO's statement that the damage from Chornobyl cannot be accurately assessed is undoubtedly true, that doesn't mean that 4,000 is a reasonable estimate of the number of cancers that will result from Chornobyl.
In "Chernobyl: Consequences of the Catastrophe for People and the Environment" published in 2009, the authors (Alexy V. Yablokov, Vassily B. Nesterenko, and Alexy V. Nesterenko with consulting editor Janette D. Sherman-Nevinger) projected that nearly a million deaths might have occurred by 2004 due to Chornobyl. Their estimates were based on reviewing 5,000 individual, small studies and small data groups, mostly studies that were NOT attempting to estimate deaths from Chornobyl, but had relevant data that could be aggregated. The Herculean task of assembling all that data produced very compelling results (one of the authors of this review meticulously reviewed the entire Chernobyl book, pre-publication, at the request of its editor, Dr. Sherman, but his edit suggestions came a few days too late to be included in the published version (he still has his hand-written notes, though!)).
That Stein and Seel reject research that doesn't align with their interests is all the more proof that more extensive research needs to be done. Or better yet, we could shut down all the reactors and spend the research dollars working on ways to sequester the waste we've already made, and cure the diseases radioactivity has already caused and exacerbated.
Sharon & Ace Hoffman, Carlsbad, California USA, January 4, 2026
The authors of the above commentary are computer programmers, and were both co-authors and programmers for a computerized interactive statistics tutorial written in the early 2000s, which was based on a book published in 1984 by Ace's father, Dr. Howard S. Hoffman, PhD, who taught statistics for nearly 50 years and was the primary author of the computer program.
Links to Dr. Ian Fairlie's analysis of INWORKS results from 2023 and 2015:
https://www.ianfairlie.org/news/important-new-bmj-article-increases-our-perception-of-radiation-risks/
https://www.ianfairlie.org/news/low-dose-radiation-is-linked-to-increased-lifetime-risk-of-heart-disease/
https://www.ianfairlie.org/news/update-new-powerful-study-shows-radiogenic-risks-of-leukemia-in-workers-more-than-double-the-previous-estimate/
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
