Coffee is a fruit that gets picked wet, dried in the open, bagged, shipped across an ocean and stored for months before anyone roasts it. Mold is not a scandal in that story — it's the default. The question that matters is what the mold left behind, and whether anyone checked. This is a plain-English guide to why mold grows on coffee, which species are the problem, what their toxins do, why roasting doesn't rescue you, and what a real lab test looks like.
Is there really mold in coffee?
Yes, routinely. Green (unroasted) coffee is an agricultural commodity grown in the tropics, and surveys of green beans from major producing countries regularly find mold on a large share of samples. That by itself is not the issue — roasting at 400°F kills living mold, and you will never see fuzz on a roasted bean. The issue is mycotoxins: the chemical by-products certain molds produce while they grow. Mycotoxins are not alive, so roasting doesn't "kill" them, and they have no taste, smell or color. A bag of beans can look, smell and brew perfectly and still carry them. That's why the only honest answer to "is this coffee clean?" is a lab report, not a sniff test.
Why coffee is such a good home for mold
Mold needs three things: a food source, warmth, and water. Coffee supplies all three at almost every stage between the tree and the roaster.
At harvest. A ripe coffee cherry is roughly half water. The bean inside has to be stripped of fruit and dried from around 50–60% moisture down to 10–12% before it's stable. In the wet (washed) process the beans sit in fermentation tanks for a day or two; in the natural process the whole cherry dries with the fruit still on. Either way the beans spend days in the exact warm, damp, sugary conditions mold loves.
During drying. Much of the world's coffee is still sun-dried on patios or raised beds. Drying that's too slow, too thin, interrupted by rain, or done directly on bare ground gives Aspergillus spores — which live in tropical soil everywhere — time to colonize the beans. Mold growth takes off when the water activity of the bean rises above about 0.80; properly dried coffee sits well below that, but it only takes a damp corner of a drying patio to create a hot spot.
In storage and transit. Dried beans are hygroscopic: they pull moisture back out of humid air. Jute bags in a warehouse during the rainy season, or a shipping container that sweats as it crosses from a hot port into a cooler one, can push the surface of the beans back into the danger zone. This is where a lot of ochratoxin contamination happens — after the farmer did everything right.
Defects and sorting. Broken, insect-damaged and over-fermented beans are the first to be colonized. Careful sorting removes most of them; cheap coffee tends to be sorted less carefully, which is one reason mycotoxin levels correlate with grade.
The molds that grow on coffee, and what they leave behind
The classic coffee molds. They thrive in warm, humid drying and storage conditions and produce ochratoxin A (OTA), the mycotoxin most associated with coffee. OTA is the reason the EU regulates coffee at all.
Better known from peanuts and corn, but they turn up on poorly stored green coffee too. They produce aflatoxins B1, B2, G1 and G2; aflatoxin B1 is one of the most potent naturally occurring carcinogens known.
A smaller group whose spores survive the temperatures that kill ordinary mold. Labs test for them separately because a product that passes a normal mold count can still carry them.
Penicillium species (another OTA producer) and Fusarium show up less often on coffee than on grains, but they're on the list of what a thorough lab looks for. The practical takeaway: a good coffee test isn't one number. It's a mold count plus a heat-resistant-mold check plus the specific toxins — OTA and the aflatoxins — because the toxins can be present when the mold that made them is long dead.
Roasting kills the mold. It doesn't remove the toxin.
Ochratoxin A is thermostable. Roasting destroys some of it — published studies disagree on how much, and it depends on roast level and method — but it does not reliably remove it, which is why regulators set limits on roasted coffee rather than just green beans. In the European Union, Regulation 2022/1370 (now consolidated into Regulation 2023/915) lowered the ceiling for OTA in roasted coffee to 3 µg/kg and in soluble (instant) coffee to 5 µg/kg, effective January 2023. Instant coffee gets a higher allowance because the extraction process concentrates whatever was in the beans — which is exactly why instant coffee drinkers should care about this more, not less.
In the United States there is no limit at all. The FDA regulates aflatoxins (20 parts per billion in human food) but has never set an action level for ochratoxin A in coffee or anything else. Nobody is checking your coffee for OTA at the border. If the brand didn't test it, it wasn't tested.
What these toxins do
The European Food Safety Authority's 2020 assessment of ochratoxin A is blunt: OTA "causes kidney toxicity in different animal species and kidney tumours in rodents" and is genotoxic in laboratory tests. EFSA declined to set a safe daily intake at all, because the way it causes kidney cancer isn't understood well enough to draw a line. The International Agency for Research on Cancer classifies OTA as a possible human carcinogen (Group 2B); aflatoxin B1 is in Group 1, a known human carcinogen, with liver damage and liver cancer as the primary concerns.
Two things make coffee a special case. First, it's a daily habit, often several times a day for decades, so even low levels add up to real chronic exposure. Second, the people who most often worry about mold in coffee — people with mold sensitivity, chronic fatigue or autoimmune conditions, pregnant women, anyone managing kidney or liver health — are also the people for whom "probably fine" isn't good enough. The honest position is that typical coffee probably contributes a small share of a typical person's OTA exposure. The equally honest position is that you can't know where your coffee falls without a number.
Why testing is the only thing that counts
Mold contamination is invisible, varies bag to bag and harvest to harvest, and survives roasting as toxin. "Mold-free" printed on a label means nothing unless there's a certificate of analysis behind it — and a surprising number of brands that say it have never published one. Here is what a real test looks like, so you can tell the difference:
A mold count (reported in colony-forming units per gram) tells you whether live mold is present in the finished product. A heat-resistant mold test catches the spores a normal count misses. An ochratoxin A assay and a total aflatoxin assay measure the toxins directly, in parts per billion, with a stated reporting limit — the smallest amount the method can reliably detect. A result of "<1 ppb" means the lab looked for it down to one part per billion and found none. The lab should be independent and named, the report should be dated and tied to a specific product, and you should be able to read it yourself.
How COFFREE is tested, and what the reports say
We send finished COFFREE sticks — the product you actually drink, not the green beans — to independent food-safety laboratories, and we publish the reports. The full panel is run by Genista Biosciences in San Jose, California, on both ends of our caffeine range: the 0% Decaf and the 100% Regular. A second lab, Eurofins Scientific, re-checked mold and yeast on the same two products in 2026 and added a microplastics screen.
| Test | 0% Decaf | 100% Regular | Lab / method |
|---|---|---|---|
| Mold | <10 CFU/g pass | <10 CFU/g pass | Genista, AOAC 2014.05 · Eurofins, FDA BAM Ch. 18 |
| Yeast | <10 CFU/g pass | <10 CFU/g pass | Eurofins, FDA BAM Ch. 18 |
| Heat-resistant mold | Negative / 25 g pass | Negative / 25 g pass | Genista, CMMEF 5th ed. Ch. 22 |
| Ochratoxin A | <1 ppb pass | <1 ppb pass | Genista, AOAC 2000.03 |
| Total aflatoxin (B1, B2, G1, G2) | <1 ppb pass | <1 ppb pass | Genista, AOAC 2005.08 |
| Moisture | 6.3% | 7.5% | Genista, AOAC 925.45 / 925.09 / 934.06 |
Genista Biosciences Certificates of Analysis 091725_172D0922R31 (Decaf) and 091725_171D0922R30 (Regular), reported September 22, 2025. Eurofins SFA report AR-26-T6-000398-01, reported July 21, 2026. "<" means below the lab's reporting limit. The Genista panel also covered lead, arsenic, cadmium and mercury (all in the low-ppb range, mercury below 1 ppb), and the Eurofins report covers microplastics across ten polymer types. Results relate to the samples submitted.
To put the ochratoxin number in context: the EU's ceiling for instant coffee is 5 ppb. Our result is below the lab's 1 ppb reporting limit — five times tighter than the strictest regulatory standard in the world, in a country that doesn't require the test at all. The same goes for aflatoxins against the FDA's 20 ppb action level. And moisture at 6–7.5% is far below the ~11–12% at which mold becomes possible again, which is what a sealed single-serve stick is for: the coffee can't rehydrate in your cupboard the way an open bag can.
Read the reports yourself.
Every claim above comes from one of these three documents. Download them, check the lab, the dates and the methods, and compare them with whatever your current coffee brand can show you.
Every COFFREE level — 0%, 25%, 50%, 75%, 100% and 150% caffeine — is the same organic, single-origin, water-decaffeinated Arabica from the same supply chain, tested the same way.
Find your caffeine level → Shop all levels →Frequently asked questions
Does roasting coffee kill mold?
Roasting kills living mold, but it does not reliably destroy the mycotoxins the mold already produced. Ochratoxin A in particular is heat-stable, which is why the EU sets limits on roasted and instant coffee, not just green beans.
What is ochratoxin A?
A mycotoxin produced by certain Aspergillus and Penicillium molds. It is the mycotoxin most associated with coffee, it is toxic to the kidneys in animal studies, and IARC classifies it as a possible human carcinogen (Group 2B).
Is there a legal limit for mold toxins in coffee in the US?
Not for ochratoxin A. The FDA has an action level for aflatoxins (20 ppb in human food) but has never set one for OTA in coffee. The EU limits OTA to 3 µg/kg in roasted coffee and 5 µg/kg in instant coffee.
Can you taste or smell mold toxins in coffee?
No. Mycotoxins are tasteless, odorless and invisible, and they survive roasting. A lab assay is the only way to know whether they're present.
Is instant coffee more likely to contain mold toxins?
Instant coffee is an extract, so whatever was in the beans is concentrated — the EU allows a higher OTA ceiling for soluble coffee for that reason. That makes testing more important for instant coffee, not less. COFFREE's instant sticks tested below 1 ppb for ochratoxin A and total aflatoxins.
What does COFFREE test for?
Mold count, yeast, heat-resistant mold, ochratoxin A, total aflatoxins (B1, B2, G1, G2), moisture and heavy metals (lead, arsenic, cadmium, mercury) by Genista Biosciences, plus a mold/yeast recheck and a 10-polymer microplastics screen by Eurofins Scientific. The reports are downloadable above.
Sources: European Food Safety Authority, "Risk assessment of ochratoxin A in food," EFSA Journal (2020); U.S. FDA, "Mycotoxins" (natural toxins in food) and CPG Sec. 555.400 aflatoxin action levels; Commission Regulation (EU) 2022/1370 and (EU) 2023/915, maximum levels of ochratoxin A in roasted and soluble coffee; International Agency for Research on Cancer monographs (ochratoxin A, Group 2B; aflatoxins, Group 1); Genista Biosciences Certificates of Analysis for COFFREE (2025); Eurofins SFA analytical report AR-26-T6-000398-01 (2026). This article is for general information and is not medical advice.