Glyphosate is one of the world’s most widely used herbicides, and it’s the active ingredient in Roundup and many other commercial weedkillers [source].

Traces of glyphosate have been detected in foods such as oats, wheat products, bread and breakfast cereals [source]. In most cases, the levels reported are below legal maximum residue limits, but its presence has raised uncomfortable questions about how farming chemicals move from fields into the food supply, and whether current safety standards are sufficient.

The controversy extends beyond laboratory measurements. It involves how glyphosate is used in farming, how much people are exposed to over time, how safety limits are set, and what the evidence can, and cannot, tell us about long-term health.

Based on current evidence, the levels typically found in food are not considered a health concern by major food-safety regulators, but that does not mean the debate is settled.

Why is glyphosate used on crops?

Glyphosate works by blocking an enzyme (EPSP synthase) involved in the shikimate pathway, a metabolic pathway found in plants and many microorganisms. Disrupting this pathway stops susceptible plants from making compounds they need to grow and survive [source, source].

In farming, glyphosate is mainly used to control weeds before a crop is planted, between crop rows, or after harvest. It can be effective, relatively inexpensive and may reduce the need for repeated cultivation. In some systems, this has helped support no-till or reduced-till farming, which can mean fewer tractor passes, lower fuel use and less soil disturbance [source].

In some countries and cropping systems, glyphosate can also be used close to harvest to control late-growing weeds or, where permitted, as a harvest aid to help crops dry more evenly. Uses and permissions vary by country, crop and product label [source, source].

Because glyphosate is applied in fields, small residues can sometimes remain on harvested crops. These can then be detected in foods made from them, including flour, bread, oats and breakfast cereals. Whether residues are present, and at what level, depends on factors including the crop, how and when the herbicide was applied, weather conditions and food processing [source, source].

This helps explain how a weedkiller can be found in food. The more difficult question is whether the amount people are exposed to creates a meaningful health risk.

Glyphosate in breakfast cereals: what testing found

In 2018, a worrying question began appearing in headlines, first in the US and then internationally: is there weedkiller in children’s breakfast cereal? And is that weedkiller linked to cancer? 

Laboratory tests commissioned by a US consumer group detected glyphosate in some popular oat-based cereals and granolas, often at parts-per-billion (ppb) levels [source]. Some samples exceeded the stricter benchmarks promoted by campaigners, including the Environmental Working Group’s health-based benchmark of 160ppb for children’s food [source]. However, most were still below official regulatory maximum residue limits (MRLs), and a campaign benchmark is not the same as a legally enforceable safety limit [source].

The image of a broad-spectrum herbicide turning up in food marketed to children helped transform a technical debate about pesticide policy into a deeply personal concern for parents. It also raised a more specific question: does the way glyphosate is used close to harvest affect the residues found in food? 

Pre-harvest use is not the same as desiccation, and permissions vary by jurisdiction. In Great Britain, for example, particular pre-harvest uses remain permitted where the product label authorises them. In the European Union, glyphosate is no longer permitted for pre-harvest desiccation, although some pre-harvest weed-control uses may still be authorised [source, source]. 

When glyphosate is applied close to harvest, there may be less time for dissipation before the grain is collected. Studies examining pre-harvest timing suggest that residues may be higher when products are applied earlier than recommended, before the recommended crop stage, or without the required pre-harvest interval [source, source].

Independent testing has detected glyphosate in some oat products and cereals, but in most cases they are below regulatory MRLs [source, source].

This helps explain how a farm chemical can end up in something like a bowl of breakfast cereal. It also explains why the debate is about more than residues alone. The questions now extend to food safety, farming practices, corporate influence and long-term health. 

Regulatory assessments generally conclude that estimated dietary intakes, including those of high consumers and children, remain below relevant safety thresholds [source, source]. Yet the headlines capture a deeper unease: not just “how much?” but “why is it there at all?” That gap between legal compliance and public expectation drives the controversy.

Three white bowls of oat hoops, granola with almonds and rolled oats on a wooden table with a jug of milk and two spoons.
Oat-based cereals and granola in breakfast bowls

Is glyphosate carcinogenic? Hazard vs risk

Much of the debate regarding glyphosate hinges on a single, often overlooked distinction: hazard vs risk. In everyday language these terms are used interchangeably, but in toxicology they answer different questions [source].

- Hazard asks: Is this substance capable of causing harm (such as cancer) under some conditions?

- Risk asks: How likely is that harm at the doses and exposure patterns people actually experience? [source]

In other words, the key issue in determining risk is not just whether glyphosate may cause cancer, but whether it is expected to do so at the concentrations encountered in real‑world use and in our diet.

In 2015 the International Agency for Research on Cancer (IARC), part of the World Health Organization, classified glyphosate as Group 2A: “probably carcinogenic to humans” [source]. This was based on limited human evidence (mainly occupational studies for non-Hodgkin lymphoma, a type of blood cancer), sufficient evidence in animals and strong supporting mechanistic data [source]. Importantly, IARC’s role is to identify hazards, but it does not assess real-world exposure levels or set safety limits. 

Regulatory agencies approach the question differently. Bodies such as the US Environmental Protection Agency (EPA) and the European Food Safety Authority (EFSA) assess risk by considering likely exposure through food, water and other routes, and comparing those exposures with toxicological reference values [source, source].

The EPA concluded in 2020 that glyphosate is “not likely to be carcinogenic to humans” at exposure levels expected when used according to approved label directions [source]. The US Ninth Circuit appeals court in June 2022 later vacated the human-health part of that decision because it found EPA had not sufficiently supported the conclusion. EPA withdrew the decision and is updating its assessment [source]. 

EFSA’s 2023 peer review similarly found “no critical areas of concern” for human health from authorised uses, while identifying some remaining data gaps [source].

The strongest human evidence on glyphosate and cancer comes from studies of people with high occupational exposure, such as farmers. These studies focus mainly on non‑Hodgkin lymphoma (NHL), the cancer most consistently raised in the debate. The Agricultural Health Study, a large long‑term cohort of US pesticide applicators, found no overall association between glyphosate use and NHL [source]. However, a 2019 meta‑analysis pooling multiple studies reported higher NHL risk in the highest exposure groups, suggesting a possible dose-response relationship at the upper end of exposure [source].

These findings do not show that ordinary dietary exposure causes cancer. They relate mainly to people who handled glyphosate-based products regularly and at much higher levels than consumers encounter through food. They may also illustrate why the evidence is debated: different studies, exposure measures and statistical approaches can lead to different conclusions, particularly when the overall association is small and exposure is difficult to measure precisely.

At first glance, these conclusions can seem contradictory. One organisation says glyphosate is “probably carcinogenic”, while others say it is not likely to pose a carcinogenic risk under authorised conditions. But they are not necessarily answering the same question.

IARC is essentially asking: “Can glyphosate cause cancer?” whereas regulators are asking: “Is the level of exposure people receive likely to cause cancer?”

Glyphosate and the gut microbiome: what the research shows

In recent years, glyphosate has attracted attention not only because of concerns about cancer, but also because of its possible effects on the human gut microbiome, part of a wider debate about pesticide residues and gut health. The concern relates to the shikimate pathway, which plants and many microbes use to make the aromatic amino acids phenylalanine, tyrosine and tryptophan. This is the biological pathway glyphosate disrupts in plants, and it is not present in human cells [source, source].

Humans need these amino acids too, but we generally obtain them from food rather than making them through the shikimate pathway [source]. Some of our gut microorganisms also possess the pathway, which could, in theory, make certain microbes potential targets of glyphosate [source]. This has prompted questions about whether real-world exposure could alter gut microbial communities or their functions in ways that might affect health.

Experimental studies show this can happen under some conditions [source, source]. In cells and animal models, glyphosate can block EPSP synthase in the shikimate pathway that some gut bacteria need to survive [source]. This may cause a build-up of certain chemicals, reduce “good” bacteria, and shift the overall balance of the gut microbiome [source]. However, findings differ between studies and may depend on dose, duration, and on whether the animals received glyphosate alone or a complete glyphosate-based formulation [source].

The gut and brain communicate closely, so these changes have been linked to signs of immune disruption and altered behaviour in animals, but other studies have found little or no meaningful behavioural effect. These findings raise the possibility that glyphosate-induced changes to the microbiome could influence immune or neurological pathways, although this remains largely based on animal and isolated cell research, and its relevance to typical human exposures is uncertain [source, source].

However, the fact that glyphosate can affect gut bacteria in a lab does not mean it’s a problem for humans at the levels people typically get from food. In the human gut, the shikimate pathway may not be very active because we usually get plenty of amino acids from our diet, so many bacteria can simply “pick up” these nutrients instead of making them themselves [source]. This could reduce the real‑world impact of partially blocking the enzyme. So, while the shikimate‑pathway mechanism is a biologically plausible reason to be cautious, current human data aren’t strong enough to say that typical dietary exposures to glyphosate cause clear, meaningful harm to gut health.

Some animal studies have used doses close to regulatory reference values, while others have used higher concentrations that may not correspond directly to ordinary dietary exposure. These differences make it difficult to compare results across studies or translate them directly to human health [source]. By contrast, biomonitoring in the general population usually finds urinary glyphosate in the microgram‑per‑litre range, consistent with much lower internal exposures [source]. This does not dismiss the findings, but it does put them in context.

Glyphosate lawsuits, the Monsanto Papers and trust

For many people, the glyphosate debate is not just about toxicology studies and regulatory reports; it is also about trust. Since 2015, more than 125,000 people have filed claims alleging that long-term exposure to glyphosate-based herbicides, including Roundup, contributed to non-Hodgkin lymphoma or other cancers. Several high-profile trials resulted in large jury awards, and Bayer agreed in 2020 to pay about $10 billion to resolve most claims pending at that time [source, source]. These cases received extensive media coverage and strongly shaped public perceptions of glyphosate, even though court verdicts are not the same as a scientific assessment of causation.

During litigation, internal company documents were disclosed and became known collectively as the “Monsanto Papers”. These were emails, memos and draft manuscripts that the plaintiffs’ lawyers said showed efforts to influence scientific debate, ghost-write articles, and lobby regulators and journalists. Critics argue that the documents reveal a pattern of corporate manipulation designed to downplay potential risks and protect sales. Industry supporters contend that some materials have been taken out of context and that the core safety conclusions remain supported by the weight of evidence [source, source].

The legal and documentary record has clearly affected trust. For campaigners and some members of the public, the “Monsanto Papers” are proof that glyphosate’s safety narrative was shaped as much by public relations as it was by independent science. For regulators and industry, the same materials are often framed as selective excerpts in an adversarial legal process, not definitive proof of harm [source, source].

This divide is important for how people interpret new findings, including those on the gut microbiome. When trust in institutions and companies is low, even tentative or preclinical results can be read as confirmation of a broader pattern of concealed risk. When trust is higher, the same studies are more likely to be seen as hypothesis‑generating, requiring much stronger human data before changing practice or policy.

A white pump sprayer with a green nozzle and a pair of green gardening gloves on a stone patio beside a lawn and weedy gravel.
An unbranded garden weedkiller sprayer beside a weedy path

What can you do about glyphosate exposure?

If you are concerned about glyphosate, there are some practical steps that you can take:

- Choose organic food: Organic standards prohibit glyphosate and other synthetic herbicides. Choosing organic versions of particular staple foods can therefore reduce exposure to glyphosate, even when conventional products comply with legal limits. However, organic food is not pesticide-free, and the nutritional and environmental advantages and trade-offs vary by product and farming system [source].

- Focus on overall diet quality: Large prospective studies and meta‑analyses consistently show that higher intakes of fruit, vegetables and whole grains are associated with substantially lower risks of cardiovascular disease, type 2 diabetes and premature death. These benefits are seen in real‑world diets that include both conventional and organic produce, and they are far larger and better established than any theoretical risk from the low pesticide residues typically found in conventional foods [source, source, source].

- Be cautious with extreme claims: Headlines about “toxic cereals” often go beyond the evidence. Look for balanced explanations that distinguish hazard from risk and acknowledge uncertainty.

Should you worry about glyphosate in food?

For most people, current evidence and regulatory assessments do not indicate that dietary exposure to glyphosate residues within permitted limits poses a significant human-health risk. However, that conclusion does not mean that every glyphosate-based product is harmless, that occupational exposure is unimportant, or that pesticide policy cannot be debated.

For most people, the science suggests that typical dietary exposures to glyphosate are unlikely to pose a large human health risk. Regulatory assessments to date have not found convincing evidence that residues in food, at levels currently authorised and monitored, cause harm to consumers, including high consumers and children. However, the topic is complex and still evolving. Legitimate questions remain about some potential effects of long-term exposure, as well as the wider impacts of farming practices on workers, communities and the environment [source, source].

It is reasonable to support farming systems that reduce unnecessary herbicide use, and to choose organic products if doing so fits your priorities and budget. At the same time, current evidence does not support alarmist claims that ordinary dietary exposure to regulated residues has been shown to cause cancer or microbiome‑related disease, and the well‑established benefits of eating plenty of fruit, vegetables and whole grains are likely to far outweigh any theoretical risk from low‑level residues in conventional produce.

A scientifically responsible position can recognise legitimate concerns about exposure, environmental effects, corporate transparency and uncertainty without implying that eating a conventional bowl of oats is, by itself, a demonstrated health danger.