The dangers of fake sugars: what the science really says

Years ago, in my early twenties, I had a strange problem. Almost every evening, after work and dinner, I would be hit by a wave of exhaustion that made no sense for someone with a desk job. On a piece of unsolicited advice, I stopped eating the sugar-free mints I chewed all day — the ones sweetened with aspartame. At the time I was sceptical. How could something approved by the FDA make me feel like that? But within two months the early-evening tiredness was simply gone, and it has never come back. That is the day I started reading ingredient labels, and the day I started distrusting the dangers of fake sugars.

This article is my attempt to separate what we actually know from the marketing. Because the biggest problem with sugar substitutes is not any single chemical — it is that they are all lumped together under one friendly word: “sweetener.” They are not the same thing, and the science treats them very differently.

The big misconception about fake sugars

When people say “artificial sweetener,” they usually mean one undifferentiated blob of chemistry. In reality there are three distinct families, and conflating them is the single most common mistake in this entire topic. The table below is the mental model I use.

Group	Examples	What it is	Main flag
Synthetic sweeteners	Aspartame, sucralose, acesulfame K, saccharin, neotame, advantame	Lab-made molecules, often hundreds of times sweeter than sugar	Cancer / DNA questions
Sugar alcohols (polyols)	Erythritol, xylitol, isomalt, maltitol, sorbitol	Carbohydrate-derived; partly absorbed; low calorie, not zero	Cardiovascular (clotting)
Natural extracts	Stevia, monk fruit	Plant-derived high-intensity sweeteners	Fewest flags, no benefit proven

This distinction matters because the World Health Organization treats them differently too. When the WHO advised against non-sugar sweeteners in 2023, it explicitly excluded sugar alcohols from that category, because polyols contain calories and are technically a different class. So before we judge any of these, we have to know which family we are talking about.

The synthetic sweeteners

This is the group I personally avoid, and the group with the most concerning recent research. Let me take the big three in turn.

Aspartame: the possible carcinogen debate

Aspartame is roughly 200 times sweeter than sugar and turns up in diet sodas, sugar-free gum, yoghurts, and — as I learned the hard way — mints. In July 2023 the International Agency for Research on Cancer (IARC), the cancer arm of the WHO, classified aspartame as possibly carcinogenic to humans, placing it in Group 2B. The basis was limited evidence for one specific cancer, hepatocellular carcinoma (a liver cancer).

Here is where it gets interesting, and where most headlines went wrong. On the very same day, the Joint FAO/WHO Expert Committee on Food Additives (JECFA) reaffirmed the acceptable daily intake of 40 mg per kilogram of body weight. In practice that means a 70 kg adult would need to drink something like 9 to 14 cans of diet soda a day to exceed it. The US FDA then rejected the carcinogen classification outright. Group 2B is the same hazard tier as aloe vera and pickled vegetables — it describes whether something could cause cancer under some conditions, not how likely it is at normal intake.

So is aspartame proven dangerous? No. But “possibly carcinogenic, with the experts admitting they could not rule out chance, bias, or confounding” is not the clean bill of health the industry implies either. For me, the calculus is simple: there is no nutritional upside to aspartame, so any non-zero question mark is enough reason to skip it.

Sucralose: the genotoxicity question

Sucralose, sold as Splenda and about 600 times sweeter than sugar, was long marketed as the inert, bulletproof option. Then a 2023 study in the Journal of Toxicology and Environmental Health from researchers at NC State and UNC Chapel Hill complicated that story. They reported that a compound called sucralose-6-acetate is genotoxic — meaning it caused DNA strand breaks in human cells in the lab. Their concern was twofold: this compound forms when the body metabolises sucralose, and trace amounts already exist in off-the-shelf product. They argued that a single sweetened drink could exceed the European safety threshold for genotoxic substances.

In fairness, this is contested. The makers of Splenda dispute that their product contains sucralose-6-acetate at detectable levels, and some toxicologists noted the metabolite clears the body within hours and questioned how well an in-vitro cell test translates to real eating. Those are legitimate objections. But “the lab signal is real and the human relevance is debated” is still a long way from “perfectly safe.” On top of that, the same line of research linked sucralose to a weakened gut lining — so-called leaky gut. Given my own focus on gut health, that alone keeps it off my list.

Acesulfame K, saccharin, neotame, and advantame

The rest of the synthetic group gets less attention but rides along on the same WHO guidance. Acesulfame K (about 200 times sweeter than sugar) and saccharin (200 to 700 times) show up constantly in processed foods and tabletop sachets. Neotame and advantame are extreme — neotame is roughly 7,000 to 13,000 times sweeter than sugar, advantame around 20,000 times — so they appear in tiny quantities. The honest position here is that the long-term human data is thin for all of them. Thin data is not proof of harm, but it is also not the reassurance the “studied for decades” marketing suggests.

Sugar alcohols are a separate story: erythritol and xylitol

Sugar alcohols like erythritol, xylitol, and isomalt are not synthetic sweeteners, and they deserve their own discussion. Two of them — erythritol and xylitol — have gone from “safe and natural” to two of the most scrutinised sweeteners on the market, and they share the same mechanism of concern.

It started with erythritol. A 2023 study in Nature Medicine found that higher blood levels of erythritol were associated with an increased risk of major adverse cardiovascular events — heart attack, stroke, and death — over a three-year window. Critics rightly pointed out two things: this was an association, not proof of cause, and erythritol is also produced naturally inside the body, so high blood levels might be an effect of metabolic problems rather than a cause. That was a fair challenge in 2023.

But the evidence kept building, and it widened to include xylitol. A follow-up intervention study showed that drinking an erythritol-sweetened beverage rapidly increased platelet reactivity in healthy volunteers — a plausible mechanism for clots — while the same dose of glucose did not. The same research group then found a near-identical pattern with xylitol, published in the European Heart Journal in 2024: a typical dietary amount sharply raised plasma levels and enhanced platelet responsiveness in every subject tested, with higher fasting xylitol levels again tied to cardiovascular events in large cohorts. A 2025 Mendelian randomization study, which uses genetics to get closer to causation, pointed in the same direction for coronary heart disease and stroke.

So yes — to answer the obvious question — xylitol carries the same cardiovascular flag as erythritol, through the same clotting mechanism. Xylitol is the more complicated case, though, because it also has a genuine, well-established benefit: it fights tooth decay, which is why it is in so much sugar-free gum and toothpaste. That local, in-the-mouth use is not the same as swallowing large amounts through processed food and drink. The dose and the route matter, and that nuance gets lost when a product simply shouts “sugar-free.”

It is worth being fair about the limitations too. Other researchers have pushed back, noting the human intervention arm was small (around ten people), had no placebo group, and measured only a single timepoint — and that elevated xylitol might be a marker of metabolic risk rather than its cause. The original team counters that their animal, cell, and whole-blood studies together point to a real causal effect. Where does that leave me? With a consistent signal across two related sweeteners, one mechanism, and one detail I cannot ignore.

That detail: commentary in the American Heart Association journals noted that older adults already have higher baseline platelet reactivity, which means this pro-clotting effect may hit us harder than the young volunteers in these studies. At 55, with cardiovascular health firmly on my radar, that is not a risk I take lightly for the sake of a slightly sweeter dessert.

The brain question: sweeteners and cognitive decline

This is the newest and, to me, the most attention-grabbing line of research — and it cuts across all three families above. In September 2025, the journal Neurology published a study of nearly 12,800 adults in Brazil, average age about 52, followed for eight years. The heaviest sweetener consumers showed roughly 62% faster decline in overall cognition — an effect the authors described as equivalent to about 1.6 extra years of brain ageing. The decline was sharpest in memory and verbal fluency, the very abilities that fade earliest in Alzheimer’s disease.

Two findings stand out. First, the association was strongest in people under 60 — not the elderly, but the middle-aged. Second, six specific sweeteners were implicated: aspartame, saccharine, acesulfame K, erythritol, sorbitol, and xylitol. One sweetener, tagatose, showed no link at all. Notice that this list mixes synthetic sweeteners and sugar alcohols — which is exactly why I treat the whole category with caution rather than trusting one “safer” subgroup.

Now the essential caveat, because this is where honesty matters most: this is observational research. It can show a relationship, not prove cause. People who drink a lot of diet products may differ in other ways — weight, diabetes, overall diet — and self-reported food data is imperfect. So on its own, the human study is a flag, not a verdict.

What makes me take it seriously is the animal work underneath it, which is the part that points at a mechanism. Animal studies suggest aspartame can trigger inflammation in the brain, while sugar alcohols like erythritol and sorbitol may alter gut microbes and weaken the blood-brain barrier. A 2024 mouse study described aspartame-driven neuroinflammation through immune cells called microglia, even at doses below the official daily limit. Most striking of all, some rodent research suggests aspartame-induced memory and learning deficits may be passed to offspring through epigenetic changes in sperm — harm that skips a generation. That is animal data, and it should not be over-read. But “human signal plus a biologically plausible animal mechanism” is a far stronger story than either piece alone.

The “natural” question mark: stevia and monk fruit

Stevia (200 to 400 times sweeter than sugar) and monk fruit extract (150 to 200 times) are the two I worry about least. They are plant-derived, they have no calories, and they do not carry the genotoxicity or cardiovascular flags above. If I had to use a sweetener, these would be my choice.

That said, “natural” is not a free pass. The WHO’s 2023 guidance against non-sugar sweeteners for weight control explicitly named stevia and its derivatives alongside the synthetics. The reason is the next section — and it applies to every sweetener in this article, natural or not.

Why “zero calories” is misleading

Here is the part the marketing never mentions. The WHO did not advise against non-sugar sweeteners because of cancer or clots. It advised against them because, after reviewing the evidence, they found that sweeteners simply do not help with weight control in the long term — and may be associated with a higher risk of type 2 diabetes and cardiovascular disease in adults.

Why would a zero-calorie product fail at weight control? The leading explanation is that sweetness without calories may decouple the taste of “sweet” from the energy your body expects to follow. Your appetite and insulin signalling were tuned over millions of years to treat sweetness as a promise of energy. Break that promise often enough and the system may compensate elsewhere. In other words, “zero calories” is an accounting fact, not a metabolic one. The body is not fooled as cleanly as the nutrition label suggests.

This is also why I think the whole framing is backwards. The real problem is our craving for intense sweetness in the first place — and that is a story about sugar itself, which I will dig into in a future Lab of One article.

A word on who funds the science

Throughout this article I have tried to give you both sides — the signal and the caveat. But I want to be honest about something that shapes how I read all of it: a lot of nutrition science is not neutral.

This is not a conspiracy theory; it is documented in the peer-reviewed literature itself. Reviews of artificial sweeteners funded by the industry were found to be roughly 17 times more likely to reach favourable conclusions than independently funded reviews. In sugar-sweetened beverage research, industry-related studies were overwhelmingly more likely to reach weak or null conclusions about harm — a pattern the authors said appeared to underestimate the health effects and may have delayed corrective public health action. When a food or drink company pays for the study, the study tends to like the product.

On top of that sits the wider replication problem: across many fields, a large share of published findings do not hold up when someone tries to repeat them. So “a study says it’s safe” is not the trump card it sounds like — especially when you do not know who paid for it.

In fairness, the bias can cut the other way too. Researchers call it “white hat bias” — the risk that independent scientists with strong convictions tilt against a product they already dislike. I keep that in mind. But here is the asymmetry that matters to me: the warning signals on fake sugars are increasingly coming from the bodies that historically gave industry the benefit of the doubt — the WHO, the IARC, independent academic groups with no product to sell. When even the cautious, consensus-bound institutions start raising flags, I stop waiting for a “final verdict” that a conflicted system may never deliver cleanly.

My approach: read the label, then decide

After that aspartame experience, I made a habit that has served me well ever since: I read the ingredient list on everything, every time. Companies reformulate products constantly, swapping sugar for sweeteners or one sweetener for another, often without any front-of-pack signal. A product you trusted last year may be a different formula today.

My personal rules are straightforward. I avoid all synthetic sweeteners — aspartame, sucralose, acesulfame K, saccharine, the lot. I treat sugar alcohols, especially erythritol and xylitol, with real caution given my age and cardiovascular focus. If I genuinely need a sweetener, I reach for stevia or monk fruit, sparingly. And above all, I do not let the word “light” or “zero” do my thinking for me. Those words are designed to sell, not to inform.

Frequently asked questions

Is aspartame actually proven to cause cancer?

No. The IARC classified it as possibly carcinogenic (Group 2B) based on limited evidence, while the WHO’s food additive committee kept the acceptable daily intake unchanged and the FDA rejected the classification. It is a question mark, not a verdict — but for a product with zero nutritional benefit, a question mark is enough for me to avoid it.

Is xylitol bad for you, even though it’s in toothpaste?

It depends on dose and route. Xylitol’s tooth-protecting benefit in gum and toothpaste is well established and not in question. The concern is swallowing larger amounts through food and drink: studies since 2024 link higher blood xylitol to increased platelet reactivity and cardiovascular events, the same mechanism flagged for erythritol. The evidence is debated, but I keep dietary xylitol low.

Do artificial sweeteners cause dementia?

There is no proof they cause dementia. A large 2025 study did link higher sweetener intake to faster cognitive decline, especially in people under 60, and animal studies suggest plausible mechanisms like brain inflammation. But the human evidence is observational, so it shows association, not cause. I treat it as a reason for caution, not alarm.

If sweeteners are zero calorie, why don’t they help with weight loss?

The WHO’s review found no long-term weight benefit from non-sugar sweeteners. The likely reason is that intense sweetness without calories disrupts the link between taste and energy that governs appetite, so the body compensates in other ways. Zero calories on the label does not mean zero metabolic effect.

Conclusion

The dangers of fake sugars are real, but they are not the same danger across the board — and that nuance is exactly what the marketing erases. Synthetic sweeteners carry the most concerning cancer and DNA questions, sugar alcohols like erythritol and xylitol raise genuine cardiovascular concerns especially as we age, and a growing body of work now links several of them to faster cognitive decline. Even the natural options offer no proven benefit for weight or health. My own position, after weighing both the evidence and the conflicts of interest behind it, is that I avoid these substances entirely — and I am not waiting for a perfect consensus to do so. My takeaway after all these years is unglamorous but reliable: eat real food, read the label every single time, and stop letting “zero” and “light” make your decisions for you. In the next Lab of One piece I will turn the lens on sugar itself — the thing all of these were supposed to replace.