Are GMOs Safe to Eat? The Rat-Tumour Study, the 51-Point Gap, and the Real Answer

Ask whether GMOs are safe to eat and you will start the most lopsided argument in science. On one side, 88% of scientists say genetically modified food is generally safe. On the other, only 37% of the public agrees — and 57% of the public says it is unsafe. That 51-point gap, measured by the Pew Research Center, is the largest disagreement between scientists and the public on any scientific issue Pew has ever polled — bigger than the gap on climate change, bigger than the gap on evolution, bigger than the gap on vaccines. One side is not merely wrong; one side is wrong by a margin that should make everyone stop and ask what is going on.

The GMO fight is really four arguments wearing one costume. The first is about safety: does eating a genetically modified plant harm you? The second is about a single study from 2012 that showed rats with tumours and still circulates as proof, a decade after it was retracted. The third is about glyphosate, the weedkiller that most GM crops are built to survive — an argument about a chemical that gets mistaken for an argument about genetics. And the fourth is about power: who owns the seeds, who profits, and whether a handful of companies should control the food supply. This article takes them in order, with the actual numbers attached, because the numbers are what settle it — and the numbers are more interesting than either slogan.

What "GMO" actually means — and what you are already eating

The first problem with the debate is the word itself. "GMO" sounds like one thing, a single laboratory trick applied to food. In practice it covers everything from a tomato engineered to ripen slowly in 1994 to rice engineered to make vitamin A, and the foods on your plate today mostly come from just a handful of crops: corn, soybeans, cotton and canola, plus sugar beets, alfalfa and papaya. More than nine in ten acres of corn, soybeans and cotton grown in the United States are genetically engineered, which means that if you have eaten corn syrup, soy sauce, cooking oil or anything made with corn starch in the last twenty years, you have almost certainly eaten genetically modified food — probably hundreds of times.

The first genetically modified food ever approved tells you how ordinary the technology's beginnings were. In 1994 the US Food and Drug Administration approved the Flavr Savr tomato, engineered to stay firm longer on the vine so it could ripen for flavour instead of being picked green and gassed red. It was not a monster; it was a tomato that stayed ripe. Four years later, genetic engineering saved an entire industry: Hawaii's papaya crop was being wiped out by the ringspot virus, and a virus-resistant papaya, developed by university researchers and given away free to farmers, rescued it. Today the Rainbow papaya is the standard papaya in Hawaii, and almost nobody who eats one knows — or cares — that it is genetically modified.

It is also worth saying what genetic modification is not. It is not injecting food with chemicals, and it is not the same as spraying a crop. A genetically engineered plant has had a specific gene added, removed or switched off — often a gene borrowed from another plant, sometimes from a bacterium — so that it gains one trait: resistance to a pest, tolerance of a herbicide, extra nutrition, longer shelf life. The plant that results is still a plant, still made of the same proteins, carbohydrates and fats, with one new protein added. That is why the safety question is answerable in the first place: regulators do not have to guess whether "GMOs" as a category are safe, because there is no such category in the body. They test each new crop, gene by gene, protein by protein, and they have been doing it for thirty years.

There is one more thing the word hides. Humans have been genetically modifying food for ten thousand years — just slowly, and blindly. Every banana you have ever eaten is a sterile mutant bred by humans until it lost its seeds. Corn is a grass called teosinte, reshaped beyond recognition by thousands of years of selective breeding. Broccoli, cauliflower, kale and Brussels sprouts are all the same species, bred into different shapes. Modern genetic engineering does in one generation, with one known gene, what traditional breeding did over centuries with thousands of unknown ones — and it does it with the gene identified, the protein tested, and the result on record. That does not make it automatically safe. It does mean the "unnatural" argument cuts both ways: almost nothing in a supermarket is natural.

What the evidence actually says: 1,783 studies and a 900-study review

The safety case does not rest on one study or one agency. It rests on a pile of evidence so large that it is hard to summarise without sounding like you are exaggerating — which is itself part of the problem, because a pile that big invites the suspicion that someone paid for all of it.

Start with the biggest single review. In 2016, the US National Academies of Sciences, Engineering, and Medicine published "Genetically Engineered Crops: Experiences and Prospects", a 388-page report compiled over two years by a committee of more than 50 scientists, researchers and agricultural experts, who reviewed more than 900 publications and heard from 80 speakers at public meetings. Its conclusion, in its own words, was that it found "no substantiated evidence of a difference in risks to human health between current commercially available genetically engineered (GE) crops and conventionally bred crops". It compared disease patterns in the United States and Canada, where GE foods have been eaten for two decades, with the United Kingdom and Western Europe, where they are barely eaten — and found no divergence: no pattern of increase in cancer, obesity, gastrointestinal illness, kidney disease, autism or allergies that tracked with GE food consumption. It also found a benefit the debate usually ignores: there was evidence that insect-resistant crops had reduced insecticide poisonings, because farmers sprayed less poison.

Two years earlier, a team led by Alessandro Nicolia published a review in Critical Reviews in Biotechnology that looked at the last ten years of GE crop safety research — 1,783 studies in total. Its conclusion: the research "has not detected any significant hazards directly connected with the use of GE crops". A separate review by Snell and colleagues in 2012 examined 24 long-term animal feeding studies — trials running from 90 days to two years, the longest feeding studies available — and found no health hazards from GE feed. The World Health Organization's position is the same in plainer language: the GM foods currently on the international market have passed safety assessments and are not likely to present risks for human health.

The sceptical reader's next question is the right one: who paid for all of this? Some of it was industry-funded, and that matters — industry-funded studies in every field tend to favour the funder's product, and the GMO literature is no exception. But the pile is not one pile. The National Academies committee was convened by an independent body, its members drawn from universities and public institutions, and its review covered independent and industry research alike. The 1,783-study review counted studies regardless of funding. And the epidemiological comparison — whole populations eating GE food for twenty years against whole populations that were not — cannot be bought, because nobody can buy a continent's cancer registry. When independent reviews, agency reviews and population-level data all point the same way, the funding objection stops being an explanation and becomes a demand for a conspiracy so large it would need to include the cancer registries of North America and Europe.

There is one honest limit to the evidence, and the National Academies stated it themselves: long-term epidemiological studies have not directly addressed GE food consumption, and subtle or long-term effects are hard to detect. Nobody has run a thirty-year randomised trial in which one group eats only GM food and the other eats none, because nobody can run a thirty-year trial of breakfast. What exists instead is twenty years of whole populations eating it, thousands of animal studies, and chemical analyses showing no meaningful differences between GE foods and their conventional counterparts. That is not proof in the mathematical sense. It is the same standard of evidence that clears every other food on the shelf — and it is a standard the anti-GMO case has never met for its own claims, which rest overwhelmingly on a single study. That study is next.

The rat study that scared the world — and why it was retracted

Every food scare has a study, and the GMO scare has one of the most famous: the Séralini study of 2012. In September 2012, French molecular biologist Gilles-Éric Séralini and his colleagues published a paper in Food and Chemical Toxicology reporting a two-year feeding study in rats. The rats fed Monsanto's Roundup-tolerant NK603 maize, or the herbicide Roundup itself, developed more tumours, suffered more kidney and liver damage, and died earlier than rats fed conventional feed. The paper came with photographs of rats with enormous tumours, released to the press under embargo — journalists were given the paper on condition they did not show it to other scientists before publication. The images went around the world in a day. Kenya banned genetically modified food imports within weeks, citing the study. The debate has never fully recovered.

Read the study itself, though, and the problems are visible from the first page. The rats were Sprague-Dawley rats, a strain that develops tumours spontaneously at very high rates — by two years of age, the majority of females develop tumours whether they eat GM maize or not. The study used ten rats of each sex per group, far too few to distinguish a real effect from chance when the background tumour rate is that high. And the results showed no dose-response relationship: rats fed the smallest amount of GM maize did not fare better than rats fed the largest amount, which is the opposite of what a real toxic effect looks like. The journal's own editor-in-chief, when he retracted the paper in November 2013, put it carefully: the results were "inconclusive" and did not reach the journal's threshold for publication, while stating unequivocally that there was no evidence of fraud or intentional misrepresentation. The authors refused to withdraw it, so the journal withdrew it without their consent.

The story did not end there. In June 2014 the same study was republished in Environmental Sciences Europe, with the data unchanged but the text revised — and without a new round of peer review; the journal's reviewers checked only that the scientific content had not changed. Its supporters still cite it as proof, and its critics still cite the retraction as proof of the opposite, and both sides are partly right about the facts and wrong about what they mean. The retraction did not prove GM maize is safe. It proved that this study could not show it was unsafe. That distinction matters, because the study is still the single most-cited piece of evidence in the anti-GMO case — and if the strongest study in your case is one that its own journal withdrew as inconclusive, your case has a problem that no amount of tumour photographs can fix.

It is also worth noting what happened after the retraction, because it shows how science actually corrects itself. The European Food Safety Authority reviewed the study and found its design, reporting and analysis inadequate, concluding that it did not allow weight to be given to its results. Multiple regulatory agencies reached the same conclusion. And in 2022, Kenya — the country that had banned GM imports in direct response to the study — lifted the ban, ten years later. The study that changed a continent's food policy did not survive its own data. The photographs, however, survive everywhere.

Glyphosate: the real argument hiding inside the GMO argument

Here is where the debate gets honest — because the strongest argument against GMOs is not about genetics at all. It is about glyphosate, the weedkiller sold as Roundup, and it is a genuinely unresolved argument between serious agencies that disagree with each other in public.

Most genetically modified crops grown today are herbicide-tolerant: they are engineered to survive being sprayed with glyphosate, so a farmer can spray a whole field and kill the weeds while the crop stands. That means the GMO debate and the glyphosate debate are the same debate in practice, and glyphosate is where the evidence stops being comfortable. In March 2015, the International Agency for Research on Cancer — part of the World Health Organization — classified glyphosate as Group 2A, "probably carcinogenic to humans", based on limited evidence in humans (studies of non-Hodgkin lymphoma in farm workers) and sufficient evidence in laboratory animals, plus strong evidence that glyphosate can damage DNA and cause oxidative stress.

But the US Environmental Protection Agency concluded in its 2017 draft assessment that glyphosate is "not likely to be carcinogenic to humans" at real-world exposures. The European Food Safety Authority found "no critical areas of concern" in July 2023, the European Chemicals Agency concluded in 2022 that glyphosate does not meet the criteria to be classified as a carcinogen, and the European Union renewed glyphosate's approval for ten years in November 2023. The same chemical, the same evidence base, opposite conclusions — and the reason is that the agencies are answering different questions. IARC rates hazard: whether a substance can cause cancer at some level of exposure. Regulators rate risk: whether it causes cancer at the levels people actually meet. Sunlight is a known carcinogen by the same hazard logic; nobody proposes banning the sun. The disagreement is real, and it is technical, and it is exactly the kind of disagreement that makes a good argument — which is why it is a pity it is usually fought under the wrong name.

Two further facts keep the argument alive. First, the courts have not settled it either: in June 2022 a US federal appeals court, the Ninth Circuit, ruled that the EPA's "not likely" conclusion was not supported by substantial evidence, because parts of the EPA's own analysis were inconsistent with it, and set aside the human-health part of the EPA's decision. The EPA says it is updating its assessment. Second, glyphosate's story has a genuine environmental cost that has nothing to do with cancer: weeds have evolved resistance to it. Glyphosate-resistant weeds — Palmer amaranth chief among them — now infest millions of acres, and farmers have responded by spraying more herbicide, and older, harsher ones, which is the opposite of what the technology promised. The National Academies report acknowledged herbicide-resistant weeds as a real and growing problem. So the honest position on glyphosate is narrower than either camp admits: the genetic modification is not the poison; the weedkiller the modification exists to tolerate is a chemical with a disputed cancer record, a real resistance problem, and a regulatory split that has not been resolved. That is a serious argument. It is just not an argument that the food itself is unsafe.

The two camps, stated fairly

Camp A — "GMOs are safe, and the opposition is anti-science." Its strongest argument is the weight of the evidence: 1,783 studies, a 900-study National Academies review, the WHO, and twenty years of whole populations eating the food with no divergence in disease patterns. It also has Golden Rice and the Rainbow papaya on its side — cases where the technology clearly did good that nothing else could do. Its weakest habit is treating "safe to eat" as the end of the argument, as if a food that will not poison you cannot still be part of a farming system worth criticising. It also tends to dismiss every critic as a conspiracy theorist, which is a good way to lose the 57% of the public who think the food is unsafe, and a bad way to win them over. And it has a funding problem it should admit more often: the companies that sell the seeds also fund much of the research, and that is a legitimate reason to demand independent verification, not a reason to be shouted down.

Camp B — "GMOs are dangerous, untested, and forced on us by corporations." Its strongest argument is the glyphosate record: a WHO agency really did call the weedkiller probably carcinogenic, resistant weeds really are spreading, and the seed business really is concentrated in a handful of companies that really do sue farmers over patents. Its labelling argument is also fair: people have a right to know what is in their food, and the fight for labelling was fought for years against industry opposition. Its weakest habit is treating a chemical, a patent system and a gene as if they were the same thing. The Séralini study is its flagship evidence, and that study was retracted as inconclusive. The "untested" claim is false on its face: no food in history has been tested more, by more agencies, for longer, than genetically modified crops. And its most common move — "we just don't know the long-term effects" — is an argument that can never be answered, because no amount of evidence can prove a negative about the next thirty years, and the same argument would forbid every new food ever introduced, including the kiwi fruit and the soybean itself.

Golden Rice: the test case that breaks both camps

If you want to see what the GMO argument is really about, look at Golden Rice — because it is the one genetically modified crop that neither camp can explain away, and it embarrasses both of them.

Golden Rice is rice engineered to produce beta-carotene, which the body converts into vitamin A. It was developed in the late 1980s and 1990s by professors Ingo Potrykus and Peter Beyer, and the technology was donated for humanitarian use — no company owns it for the farmers who need it. The need is not abstract: vitamin A deficiency is a major cause of preventable blindness in children, with hundreds of thousands of children going blind every year and an estimated 190 million children worldwide affected. In the Philippines, where rice is the staple food, only about two in ten households meet their estimated vitamin A requirement, and roughly 15.5% of children under five are vitamin A deficient. A single cup of cooked Golden Rice can provide 30% to 50% of a young child's estimated average requirement for vitamin A. It is, by any honest measure, a public-health intervention that happens to be a plant.

And yet Golden Rice spent more than twenty years fighting for permission to exist. Field trials were destroyed by activists in the Philippines in 2013. Greenpeace campaigned against it for decades, arguing that the answer to vitamin A deficiency was dietary diversity and supplementation, not engineered rice — a fair argument in a rich country, and a crueller one in a place where the poorest families eat rice three times a day because that is what they can afford. In June 2016, 107 Nobel laureates signed an open letter urging Greenpeace to stop opposing genetically modified crops, and Golden Rice in particular, calling the opposition a "crime against humanity". The Philippines finally approved Golden Rice for commercial propagation in July 2021 — the first country in the world to do so — and pilot plantings followed in 2022 and 2023, with 97% of the target beneficiaries surveyed saying they wanted it in their diet.

Golden Rice breaks Camp B's case, because it is a GMO developed by public scientists, given away free, that prevents children going blind — and opposing it on "corporate control" grounds requires pretending a donated public-health crop is a Monsanto product. And it breaks Camp A's case too, because the twenty-year delay proves that "the science is settled" was never the whole story: the technology was ready and the evidence was strong, and it still took a generation to reach the children it was built for, because the argument was never only about evidence. It was about trust, and trust is not won by citing 1,783 studies at people who have watched corporations sue farmers and regulators disagree about a weedkiller in public. Golden Rice is the proof that GMOs can do unambiguous good, and the proof that the GMO debate was lost, for decades, on ground that had nothing to do with safety.

What the critics get right

A verdict for the food is not a verdict for the system, and the critics of GMOs are right about more than their opponents admit. The first is corporate control. The seed business is concentrated in a handful of companies, and the patent system gives them real power over farmers: in 2013 the US Supreme Court ruled unanimously in Bowman v. Monsanto that a farmer could not replant second-generation seeds from a patented crop without paying again. You can think the food is safe and still think that is a bad way to organise the food supply. The second is monoculture. Herbicide-tolerant crops made it easy to plant the same crop, in the same field, year after year, sprayed with the same chemical — and that is how you breed resistant weeds and strip the diversity out of a landscape. The third is the labelling fight: for years, the industry resisted labelling GM foods, and it took a federal disclosure standard in the United States, in force since 2019, to require "bioengineered" labelling at all. People have a right to know what they are eating, whatever the safety evidence says. The fourth is the environment, where the National Academies' own report is more cautious than the safety verdict: it found no conclusive evidence of environmental harm, but also found real trade-offs — insect-resistant crops reduced insecticide use, herbicide-tolerant crops increased herbicide use, and gene flow to wild relatives is a genuine management problem, not a fantasy.

None of this makes the food unsafe. But it explains why the argument will not die: because the people arguing about safety are often, without knowing it, arguing about power, and the people answering about safety are often, without knowing it, defending power. The way out is to separate the questions. Is the food safe to eat? The evidence says yes, and the evidence is large. Is the farming system built around it good? That is a real argument, and the critics have real points. Is glyphosate safe? That is genuinely disputed, and the dispute deserves its own honest answer instead of being smuggled into a debate about genes. A debate that asks one question at a time is a debate that can end. A debate that asks four questions under one word — GMO — will run for another thirty years.

At a glance: claim vs evidence

Claim you will hear	What the evidence actually says
GMOs are untested	False. No food in history has been tested more: 1,783 studies in one 2014 review, 900+ in the 2016 National Academies report, plus 30 years of regulatory testing crop by crop.
The 2012 rat study proves GMOs cause cancer	It does not. The Séralini study used 10 rats per group of a tumour-prone strain, showed no dose-response, and was retracted in 2013 as inconclusive. It was republished in 2014 without new peer review.
Scientists agree GMOs are safe	Yes, overwhelmingly. 88% of AAAS scientists say GM food is generally safe to eat. But only 37% of the public agrees — a 51-point gap, the largest Pew has measured.
Glyphosate causes cancer	Genuinely disputed. IARC (WHO) calls it "probably carcinogenic" (2015); the EPA says "not likely" at real-world exposure; EFSA found no critical concerns (2023). The split is hazard versus risk — and a US appeals court vacated the EPA's conclusion in 2022.
Nobody knows the long-term effects	Partly true and partly beside the point. No 30-year trial exists. But 20+ years of whole populations eating GE food shows no divergence in disease patterns versus Europe, and 24 long-term animal feeding studies found no hazards.
GMOs are just a corporate money grab	The seed business is concentrated and patent power is real — the Supreme Court backed Monsanto over replanted seed in 2013. But Golden Rice was developed by public scientists and donated free, and the Rainbow papaya was given free to Hawaiian farmers.
GMOs will destroy the environment	Mixed, and the honest answer is mixed. Insect-resistant crops cut insecticide spraying and poisonings; herbicide-tolerant crops increased herbicide use and bred resistant weeds. The National Academies found no conclusive evidence of environmental harm, but real trade-offs.
"Non-GMO" on the label means healthier	No. The label tells you about the breeding method, not the nutrition. A non-GMO biscuit is still a biscuit. The US requires "bioengineered" disclosure; the EU requires GMO labelling.
Golden Rice proves GMOs save lives	It proves the technology can do real public-health good: a cup provides 30–50% of a young child's vitamin A requirement. It also proves the debate was never only about evidence — it took 20+ years to reach approval despite the data.

Popular opinions: how divided people actually are

The arguments below are paraphrased from the recurring positions in forums, social media threads and comment sections — not quoted posts, and no usernames are invented or cited. They show the real shape of the divide, which is not "science versus ignorance" but different answers to different questions.

"The rats had tumours the size of golf balls — how can you say it's safe?" The photograph camp, and the most shared argument in the debate. Its strength is that the images are real photographs from a real published study, and they are horrifying. Its weakness is that the study behind them used ten rats per group of a strain that grows tumours anyway, showed no dose-response, and was retracted by its own journal as inconclusive. Verdict: a true photograph from a study that could not support its own conclusion — evidence of a bad study, not of a dangerous food.

"Scientists say it's safe, but scientists said cigarettes were safe too." The trust camp, and the one that explains the 51-point gap better than any other. Its strength is historical: industries have captured science before, and the seed companies do fund research. Its weakness is that it proves too much — the same logic would dismiss the safety of every approved medicine, every vaccine and every food additive, and the GMO evidence includes independent reviews, agency reviews and population-level data that no company could fabricate. Verdict: a legitimate reason to demand independent verification, not a reason to dismiss the verification when it arrives.

"It's not the gene, it's the Roundup." The most accurate slogan in the debate, and the one the evidence supports most directly. Its strength is the glyphosate record: a WHO agency really did call it probably carcinogenic, resistant weeds really are spreading, and the chemical is the reason most GM crops exist. Its weakness is that it is usually deployed as a reason to ban the crop rather than to regulate the chemical — as if the answer to a disputed weedkiller were to ban vitamin A rice. Verdict: the right argument, aimed at the wrong target.

"We've been eating it for twenty years and we're fine." The lived-experience camp, and the most common defence online. Its strength is the population data: the United States and Canada have eaten GE food for two decades, Europe has barely eaten it, and the disease patterns did not diverge. Its weakness is that "we're fine" is doing a lot of work — obesity, diabetes and cancer all rose over the same twenty years, and the claim only works because the rise happened on both sides of the Atlantic. Verdict: a true statement about populations, not a proof about individuals, and useless against someone who distrusts the population data itself.

"Monsanto sued farmers for saving seeds — that's all you need to know." The power camp, and the argument that actually wins converts. Its strength is that it is true, and that the Bowman v. Monsanto decision is a matter of public record. Its weakness is that it is an argument about patents, not about safety — and it proves nothing about whether the food is safe to eat, only about who controls it. Verdict: the strongest argument in the debate, and it is not an argument about safety at all.

"If it's so safe, why did they fight labelling for years?" The transparency camp, and the question the industry has never answered well. Its strength is that the labelling fight really happened, and the industry really did resist it, which is strange behaviour for a product with nothing to hide. Its weakness is that it confuses a marketing fight with a safety finding — companies fight labelling for all kinds of reasons, including the reasonable fear that a label will be read as a warning. Verdict: a fair question about trust, and the industry's answer to it damaged its own case more than any study ever did.

Notice the pattern, familiar from every food fight: each side reads the same facts as proof of its own position. One side hears "no evidence of harm" and stops listening. The other hears "probably carcinogenic" and stops listening. The official position — the food is as safe as its conventional counterpart, the weedkiller is genuinely disputed, and the seed business deserves scrutiny — is less satisfying than either slogan, because it requires you to hold three different verdicts at once instead of one. That is not evasion. It is the answer.

How to decide for yourself

A practical approach that follows the evidence, rather than a camp, looks like this. If the question is whether the genetically modified food on the shelf is safe to eat, the answer is yes by every large review ever conducted — eat it, or don't, for reasons of taste, price or principle, but not because you believe it will poison you. If the question is whether to worry about glyphosate, the honest answer is that the agencies disagree and the science is genuinely unresolved at the margins: washing produce, eating a varied diet, and favouring organic for the foods you eat most often are all reasonable precautions that cost little. If the question is whether to support the system, that is a political choice, not a scientific one: buy organic or local if you want to vote against concentrated seed ownership, and know that you are voting on power, not on safety. And if you want a single fact to carry away from the whole argument, make it this one: the most important genetically modified crop in the world is a bowl of rice that stops children going blind, and it took twenty years of argument to reach them — not because the science was unclear, but because the argument was never really about the science.

The GMO fight will not end with this article, and it should not. The glyphosate question is still open. The patent question is still open. The monoculture question is still open. What is closed, as closed as a question like this ever gets, is the safety of the food itself: 1,783 studies, 900 more in the National Academies review, twenty years of whole populations eating it, and no evidence of harm that has survived examination. The 51-point gap between scientists and the public is not a gap in the evidence. It is a gap in trust — and trust is not rebuilt by another review. It is rebuilt by labelling honestly, by regulating the weedkiller on its own merits, by letting public-interest crops like Golden Rice reach the people they were built for, and by admitting, out loud, that the critics were right about the things they were right about. The food is safe. The argument was never only about the food.

Frequently asked questions

Are GMOs safe to eat?

Yes, according to every large review ever conducted. The US National Academies reviewed more than 900 studies and found no substantiated evidence of a difference in health risks between today's genetically engineered crops and conventionally bred ones. A review of 1,783 studies reached the same conclusion, as did the World Health Organization. That verdict covers the GM foods currently on the market — each new crop is tested individually.

What does GMO mean?

A genetically modified organism has had a specific gene added, removed or switched off in the laboratory — for example, a gene that makes a crop resistant to a pest or tolerant of a herbicide. Most GM food comes from corn, soybeans, cotton and canola. More than nine in ten acres of US corn, soybeans and cotton are genetically engineered, so most people in North America have eaten GM food many times.

Did the Séralini rat study prove GMOs cause cancer?

No. The 2012 study reported more tumours in rats fed GM maize, but it used only ten rats per group of a strain that develops tumours spontaneously, showed no dose-response relationship, and was retracted by its journal in 2013 as inconclusive. It was republished in 2014 without new peer review. Regulators including the European Food Safety Authority found its design inadequate.

Is glyphosate dangerous?

Genuinely disputed. The WHO's cancer agency IARC classified glyphosate as "probably carcinogenic" in 2015, while the US EPA said it is "not likely" to cause cancer at real-world exposures and European regulators found no critical concerns in 2023. The split is partly hazard versus risk: IARC asks whether it can cause cancer at any exposure, regulators ask whether it does at the levels people actually meet. Glyphosate-resistant weeds are a separate, real and growing problem.

Why do so many people think GMOs are unsafe if scientists disagree?

The Pew Research Center found 88% of scientists say GM food is generally safe, but only 37% of the public agrees — a 51-point gap, the largest scientist–public split Pew has measured. The gap is driven by distrust of the companies involved, the retracted rat study's tumour photographs, the glyphosate dispute, and the industry's long fight against labelling — not by the safety evidence itself.

What is Golden Rice?

Golden Rice is rice engineered to produce beta-carotene, which the body converts into vitamin A. Developed by public scientists and donated for humanitarian use, it can provide 30–50% of a young child's vitamin A requirement in a single cup. The Philippines approved it for commercial growing in July 2021, the first country to do so, after more than twenty years of opposition — including a 2016 open letter from 107 Nobel laureates urging Greenpeace to stop blocking it.

Sources and further reading

• National Academies of Sciences, Engineering, and Medicine, "Genetically Engineered Crops: Experiences and Prospects", 2016 — the 388-page consensus report reviewing 900+ publications; "no substantiated evidence of a difference in risks to human health between current commercially available genetically engineered (GE) crops and conventionally bred crops". https://nap.nationalacademies.org/catalog/23395/genetically-engineered-crops-experiences-and-prospects
• Nicolia A et al., "An overview of the last 10 years of genetically engineered crop safety research", Critical Reviews in Biotechnology, 2014;34(1):77–88 — review of 1,783 studies finding no significant hazards directly connected with GE crops.
• Snell C et al., "Assessment of the health impact of GM plant diets in long-term and multigenerational animal feeding trials: a literature review", Food and Chemical Toxicology, 2012;50(3–4):1134–1148 — 24 long-term feeding studies, no health hazards found.
• Pew Research Center, "Public and Scientists' Views on Science and Society", January 2015 — 88% of AAAS scientists vs 37% of the public say GM foods are generally safe; 57% of the public says unsafe; a 51-point gap. https://www.pewresearch.org/science/2015/01/29/chapter-3-attitudes-and-beliefs-on-science-and-technology-topics/
• Séralini GE et al., "Long term toxicity of a Roundup herbicide and a Roundup-tolerant genetically modified maize", Food and Chemical Toxicology, 2012;50(11):4221–4231 — retracted November 2013 as inconclusive; republished in Environmental Sciences Europe, 2014;26:14, without new peer review.
• National Academies of Sciences, Engineering, and Medicine, Box 5-5, "Controversial Results of an Animal Feeding Study of Genetically Engineered Crops and Glyphosate", in Genetically Engineered Crops: Experiences and Prospects, 2016 — the retraction history and the committee's assessment. https://www.ncbi.nlm.nih.gov/books/NBK424534/box/box_5_5/
• International Agency for Research on Cancer, IARC Monographs Volume 112, 2015 — glyphosate classified Group 2A, "probably carcinogenic to humans", on limited human evidence and sufficient animal evidence.
• European Food Safety Authority, "Glyphosate: no critical areas of concern; data gaps identified", July 2023 — the European regulator's assessment, and the basis for the EU's ten-year renewal in November 2023.
• USDA Foreign Agricultural Service, "Philippines First in World to Approve Golden Rice for Propagation", July 2021 — the July 21, 2021 approval, and the vitamin A deficiency figures behind it.
• World Health Organization, "Food, genetically modified" — GM foods currently on the international market have passed safety assessments and are not likely to present risks for human health. https://www.who.int/news-room/questions-and-answers/item/food-genetically-modified
• Bowman v. Monsanto Co., 569 U.S. 278 (2013) — the unanimous Supreme Court decision that replanting second-generation patented seed infringes the patent.

This article is for general information only. If you have a specific health concern about any food, speak to a clinician or a qualified nutrition professional about your personal situation.

Found this answer useful? This post is part of a series that takes questions many people search for but nobody has answered properly, and gives the real, detailed answer. Earlier in the series: whether eggs are bad for your cholesterol, whether vaping is safer than smoking, whether aspartame is safe, whether sugar makes kids hyperactive, whether red meat causes cancer, whether fluoride in tap water is safe, whether seed oils are toxic, whether organic food is actually healthier, the MSG food fight and whether expired poison becomes safe. If you have a question you want answered this way, reach out through asadwebs.com.