Artificial Sweeteners in 2026: The Aspartame Verdict and What Came After
IARC called aspartame a possible carcinogen. WHO recommended against sweeteners for weight loss. Here's where the science stands now.
July 2023 was a watershed moment for artificial sweeteners. The International Agency for Research on Cancer (IARC) classified aspartame as “possibly carcinogenic to humans” — Group 2B. Headlines erupted. Diet soda drinkers panicked. The food industry pushed back hard.
Then, the same month, the Joint FAO/WHO Expert Committee on Food Additives (JECFA) reaffirmed its existing acceptable daily intake for aspartame at 40 mg/kg body weight per day. Two bodies under the WHO umbrella. Two seemingly contradictory conclusions about the same molecule.
The confusion was understandable. But what happened next — a quieter WHO guideline, a wave of microbiome research, and a reckoning with “natural” alternatives — reshaped the sweetener landscape in ways that the aspartame headlines never could.
The IARC Classification Explained
IARC’s Group 2B designation means “limited evidence of carcinogenicity in humans.” That sounds alarming until you look at what else sits in Group 2B: aloe vera extract, pickled vegetables, talc-based body powder, and radiofrequency electromagnetic fields (the kind your phone emits). The category is crowded with substances that have shown some signal in some studies but lack the consistency or strength of evidence to move higher.
The critical distinction that most coverage missed: IARC performs hazard identification, not risk assessment. Hazard identification asks can this substance cause cancer under any conditions, at any dose? Risk assessment asks will it cause cancer at the doses people actually consume? These are fundamentally different questions. Sunlight is a Group 1 carcinogen (the highest category), but no one argues you should never go outside.
The evidence IARC relied on came primarily from three large cohort studies that found associations between aspartame consumption and hepatocellular carcinoma (liver cancer). The French NutriNet-Santé cohort (Debras et al., 2022, PLOS Medicine) was the most cited, reporting a modest increase in cancer risk among higher consumers. But the studies had confounding issues that the researchers themselves acknowledged: people who consume more artificial sweeteners tend to differ from non-consumers in ways that are difficult to fully control for — they are more likely to be overweight, more likely to be managing diabetes, and more likely to have other dietary patterns that independently affect cancer risk.
JECFA, looking at the same body of evidence through a risk assessment lens, concluded that the data did not provide convincing evidence of harm at typical consumption levels. A person weighing 70 kg would need to drink roughly 9 to 14 cans of diet soda daily to exceed the ADI. Most people consume a fraction of that.
Both bodies did their jobs correctly. They were just answering different questions.
The WHO Guideline That Mattered More
Six weeks before the IARC classification grabbed headlines, the WHO released a guideline that arguably mattered more for public health. In May 2023, the WHO formally recommended against the use of non-sugar sweeteners for weight control.
This was not about cancer. It was about efficacy. A systematic review of 283 studies found that non-sugar sweeteners showed no long-term benefit for reducing body fat or preventing weight-related diseases. Short-term studies sometimes showed modest calorie reduction, but anything longer than a few months showed the effect vanishing. Some longer-term observational data even suggested associations between regular sweetener use and increased risk of type 2 diabetes and cardiovascular disease — though these associations could reflect reverse causation (people already at risk choosing diet products).
This guideline questioned the entire premise of diet drinks. The implicit promise of artificial sweeteners has always been simple: same taste, fewer calories, therefore less weight gain. The WHO looked at decades of evidence and said the “therefore” does not hold up. The reasons are still debated — compensatory eating, disrupted appetite signaling, metabolic effects — but the outcome data are stubbornly consistent.
The food industry responded that the WHO was conflating different sweeteners with different metabolic profiles, which is a fair point. Aspartame, sucralose, stevia, and sugar alcohols behave differently in the body. Grouping them as “non-sugar sweeteners” obscures those differences. But the WHO’s conclusion was about the category-level claim that substituting sweeteners for sugar leads to lasting weight management, and at that level, the evidence was not supportive.
Sucralose: The Gut Microbiome Disruption
While aspartame dominated the cancer conversation, sucralose was quietly accumulating a different kind of evidence — not about carcinogenicity, but about biological activity in the gut.
The landmark study came from Suez et al. (2022, Cell). In a randomized controlled trial of 120 participants, researchers found that sucralose and saccharin altered glycemic responses through changes in the gut microbiome. The effects were personalized — not everyone responded the same way — but in those who did respond, the changes were measurable within two weeks. The microbiome composition shifted, and those shifts correlated with impaired glucose tolerance.
This was not a petri dish study. It was a well-designed human trial published in one of the most rigorous journals in biology. And it built on earlier work. Schiffman and Rother (2013, Journal of Toxicology and Environmental Health) had already documented that sucralose affects P-glycoprotein, a transporter protein in the intestinal wall that plays a key role in drug absorption. Sucralose, at dietary doses, was shown to inhibit this transporter — meaning it could theoretically alter how the body processes certain medications.
None of this means sucralose is “toxic.” It means that the long-held assumption that non-caloric sweeteners pass through the body inertly, doing nothing except tasting sweet, is no longer tenable. Sucralose is biologically active at doses people actually consume. Whether those effects translate to meaningful health consequences over years of daily consumption is an open question, but “biologically inert” is off the table.
More recent work has continued in this direction. Multiple research groups have now documented microbiome alterations from various artificial sweeteners in human subjects, though the magnitude and clinical significance of these changes remain subjects of active investigation. The gut microbiome field itself is still young, and our ability to say what constitutes a “good” or “bad” microbiome composition is limited.
Stevia and Monk Fruit: The “Natural” Alternatives
As confidence in traditional artificial sweeteners has eroded, stevia and monk fruit have surged in popularity. They are marketed as natural alternatives — plant-derived, zero-calorie, and free from the baggage of synthetic chemistry.
The marketing, however, elides some important details. The stevia you find in packets and ingredient lists is not stevia leaf. It is steviol glycosides — highly purified chemical compounds extracted from the stevia plant through a multi-step industrial process involving ethanol extraction, resin purification, and crystallization. The FDA has approved high-purity steviol glycosides (Reb A, Reb M, and others) as generally recognized as safe, but has not approved crude stevia leaf extracts. The purity specification exists for a reason: whole-leaf stevia contains compounds that raised reproductive toxicity concerns in older animal studies.
Monk fruit follows a similar pattern. The sweetness comes from mogrosides — triterpene glycosides that undergo industrial extraction and purification. The end product is as processed as any other food additive. Calling it “natural” is a regulatory and marketing designation, not a description of how it reaches your tongue.
Both sweeteners are generally well-tolerated, and their safety profiles are genuinely less concerning than those of aspartame or sucralose based on current evidence. But they are not without open questions. A 2019 in vitro study (Denina et al., Letters in Applied Microbiology) found that steviol glycosides could inhibit certain strains of gut bacteria at concentrations relevant to dietary intake. In vitro findings do not automatically translate to in vivo effects, but given what we have learned about sucralose and the microbiome, dismissing the possibility would be premature.
The bigger concern with “natural” sweeteners may be psychological rather than biochemical. The health halo effect — the tendency to assume that natural means safe and safe means beneficial — leads consumers to treat these products as if they have no downsides at all. That assumption has not been earned by the evidence. It has been earned by the marketing.
The Sugar Alcohol Middle Ground
Sugar alcohols — erythritol, xylitol, sorbitol, and their relatives — occupy an interesting middle ground. They are not non-caloric (most provide 1.5 to 3 calories per gram, compared to sugar’s 4), they are not as sweet as high-intensity sweeteners, and they have a different set of tradeoffs entirely.
Erythritol had been considered one of the safer options until February 2023, when a Cleveland Clinic study (Witkowski et al., Nature Medicine) reported an association between blood erythritol levels and increased risk of major adverse cardiovascular events including heart attack and stroke. The study found that erythritol enhanced platelet reactivity and thrombus formation in vitro and in animal models.
The study made waves, but important caveats apply. The observational component showed correlation, not causation. The participants in the high-erythritol group were already at elevated cardiovascular risk. And the body produces erythritol endogenously — high blood levels might be a marker of metabolic dysfunction rather than a cause. Follow-up research is ongoing, and neither the FDA nor EFSA has changed erythritol’s regulatory status based on this single study. But it was a reminder that “well-tolerated” and “biologically inconsequential” are not the same thing.
Xylitol, widely used in sugar-free gum and dental products, has well-established dental benefits — it genuinely reduces cavity-causing bacteria. It also has a well-established veterinary risk: xylitol is severely toxic to dogs, causing rapid insulin release that can lead to hypoglycemia, liver failure, and death. Households with dogs need to treat xylitol-containing products with the same care as chocolate.
Sorbitol, found naturally in stone fruits and used widely in sugar-free candy, has the most predictable side effect profile: it is an osmotic laxative. At moderate doses (above roughly 10 grams), it draws water into the intestine and causes gastrointestinal distress. The “sugar-free gummy bear” reviews that became an internet phenomenon were documenting a well-characterized pharmacological effect, not an anomaly.
No Free Lunch — or Free Sweetness
Three years after the IARC classification, the sweetener landscape looks less like a simple hierarchy of safe-to-dangerous and more like a set of tradeoffs, each with its own evidence base and uncertainty profile.
Sugar causes metabolic disease at the doses many people consume. Aspartame has a possible-but-unproven cancer signal and a solid safety record at typical intake. Sucralose is biologically active in the gut in ways we are still characterizing. Stevia and monk fruit are well-tolerated but less studied and not as “natural” as their labels suggest. Sugar alcohols have their own cardiovascular questions and gastrointestinal realities.
No sweetener is consequence-free — including the one we have used for thousands of years. The question is not which option is “safe” in some absolute sense, but which tradeoffs align with your own health priorities. GradeMyLabel scores each sweetener individually based on its regulatory status, risk evidence, and study data, so you can see exactly where the science stands for each one on every label you scan.