Girl Scout cookies are safe to eat, scientists confirm

Everybody knows Girl Scout cookies contain fat and sugar, because, well, they’re cookies. But an analysis released in December, commissioned by consumer advocacy nonprofits GMOScience and Moms Across America, makes a bolder claim.

In a self-published article, the groups contend that Girl Scout cookies contain an “extremely concerning” level of “toxic contaminants”—including glyphosate, a widely used pesticide; aluminum; mercury; cadmium; and lead. First published in December 2024, the article made headlines when Joe Rogan mentioned the report on his podcast in February.

All the study authors “were Girl Scouts when we were children,” says report coauthor Stephanie Seneff, a senior research scientist at the Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory and author of Toxic Legacy: How the Weedkiller Glyphosate Is Destroying Our Health and the Environment.

“So we felt that the Girl Scouts had a potential opportunity to take on the task of challenging the government to say that there's too many toxins in our food, and we need to fix it.”

But as it turns out, the case against the Girl Scout cookies is thinner than a Thin Mint.

“If we look at the numbers [of heavy metals and pesticides] that were published, those numbers are really typical. They’re what we see in food,” says Meghan Cahill, an analytical chemist at the Connecticut Agricultural Experiment Station who specializes in the elemental analysis of food and whose opinions are not those of her employer.

In fact, the numbers contained in the advocacy groups’ independent analysis—which was not submitted to a scientific journal for peer review—were so low, Cahill says she feels better about eating Girl Scout cookies than she did before.

“I’m a Tagalong person,” Cahill says as she pulls a box of Girl Scout cookies into the frame of the Zoom call and takes a bite of cookie.

The dose makes the poison

Norbert Kaminski, a toxicologist and director of Michigan State University’s Center for Research on Ingredient Safety, agrees with Cahill.

“All the levels that they measured, the metals, as well as the glyphosate, they’re well below the [US Food and Drug Administration] limits.”

To be clear, no one is saying that Girl Scout cookies contain zero heavy metals, which occur naturally in trace amounts in Earth’s crust and can leach into wheat and other vegetables. It’s only that the levels are too low to harm someone—points that the Girl Scouts of the USA noted on their website in February.

That’s because, as they say in toxicology, the dose makes the poison.

For instance, a person would have to consume 92 Girl Scout cookies each day for the rest of their life to reach a harmful level of aluminum—a concept called the reference dose, Kaminski says. He and his colleagues identified the highest identified level of each contaminant in the GMOScience study and then used existing food-safety standards to calculate what would constitute a safe dose. Kaminski and his team published a blog post about their findings in March.

Kaminski’s group calculates that a child could eat 412 cookies a day before reaching a harmful dose of mercury. Concerns over lead ingestion might kick in at a rate of four cookies eaten each day for some of the cookies tested in the GMOScience analysis—but again, this means eating four cookies each day, every day, for the rest of your life.

While no level of lead is considered safe, the FDA also recognizes that completely removing it from food is unrealistic, which is why it uses 2.2 µg/day as an interim reference level for children. Exposure to high amounts of lead in childhood can cause lifelong neurological and developmental issues.

Glyphosate, a synthetic chemical best known as the active ingredient in some herbicides, notably Roundup, also exists in tiny amounts in our food. Despite the controversy surrounding the substance, US and European Union regulators have judged that it is not a carcinogen. To approach its reference dose, you’d have to eat 74,000 cookies each day for a lifetime, Kaminski says.

Yet the threshold for what is considered safe to eat or drink depends on whose set of numbers you go by, says John Fagan, chief scientist and CEO of Health Research Institute, a nonprofit laboratory based in Iowa that sampled the 25 cookies for the GMOScience report.

“The level of glyphosate that the [US Environmental Protection Agency] claims is safe on a daily basis for the rest of your life is four times the level that the European Food [Safety] Agency says is safe,” Fagan says. The EPA threshold is 50 times as high as what California law considers acceptable.

Even using the most conservative numbers cited by Fagan gives an unattainable tally. A child could eat 370 Girl Scout cookies a day for a lifetime before exceeding what California deems a dangerous dose of glyphosate.

“The take-home message is that the analytical methods we have are exquisite in terms of being able to detect” substances such as glyphosate, Kaminski says. “But just because you can detect something doesn't mean that it's going to cause harm.”

Wrong comparison

Both Cahill and Kaminski take issue with the fact that the GMOScience analysis compared the levels of metals and glyphosate found in the cookies to the EPA’s limits for levels in water.

This is because the water limits are based on safety levels for an adult male individual weighing approximately 175 pounds (about 80 kg) and consuming about 2 L of water each day.

Of course, Kaminski points out, no one is likely to consume the number of cookies equal to the weight of a 2 L bottle of water daily.

“The comparison is totally inappropriate,” he says.

Seneff says her team used the water limits because they couldn’t find any good limits for heavy metals in food. “I don’t think there are levels that are specified,” she says.

Kaminski counters that not only do these limits exist, but they can be accessed by anyone on the EPA’s website. The reference dose for cadmium, for example, is well established: 0.001 mg/kg/day. And at that rate, an adult male could eat an estimated 52 Girl Scout cookies each day for a lifetime before experiencing adverse effects.

When told of Kaminski’s calculations, Seneff said they were “good news.”

“I was actually quite concerned about whether we were going out too far in claiming the metals were a problem,” Seneff says. “I was kind of uneasy about that whole aspect of it.”

Missing details

Seneff and colleagues’ sample size of 25 cookies is not sufficient to draw conclusions about the 200 million boxes of cookies the Girl Scouts distribute annually across the US, Cahill and Kaminski say.

“This is not representative of the product. This is representative of whatever they tested. And we don't know how they tested it because they didn't tell us how they tested it,” Cahill says.

Disclosing the research methodology is standard in any rigorous study, yet it was missing from the advocacy groups’ research article. So were confidence intervals, which explain how much room for error there might be.

“I can say that, for heavy metal analysis, my lab operates with a 10% method uncertainty,” Cahill says.

For his part, Fagan says that his lab can produce both methodology and confidence interval standards, should they need to do so in court. Danielle Barbaro and Judy Cholewa, both of New York City, are suing Girl Scouts of the United States of America, as well as its cookie manufacturers, Little Brownie Bakers of Ferrero USA and ABC Bakers of Interbake Foods, for selling contaminated cookies that they claim violate New York consumer protection laws. “We’re very disappointed about that,” Seneff says. “We certainly don’t endorse the lawsuit.”

When asked if she and her colleagues plan to publish confidence intervals, methodology, or any more related data on Girl Scout cookies, Seneff says, “I don't think we're planning on doing anything else with this study.”

Jason Bittel is a science writer based in Pennsylvania.