"Anti-sugar warnings are just 'food anxiety,' and artificial sweeteners like aspartame have no health risks."
That sentence is a compression of a message that spread across TikTok and Instagram in 2023, not a quotation: none of the health professionals in our sources is on record saying "no health risks", and that distinction decides the grade.
Two arguments are bundled here — that warnings about sugar are a psychological problem rather than a nutritional one, and that the sweeteners used instead carry no risk. The first is contradicted. The second is false as an absolute, though its sturdier version, that aspartame is not a meaningful safety concern for most people at ordinary intakes, is supported by every regulator we consulted.
Part of this message was paid for
The aspartame half has a documented origin. A joint investigation by The Examination and The
Washington Post, published on 13 September 2023, reviewed more than 6,000 posts from 68
registered dietitians; of the 33 posting paid content, roughly 17 "did not clearly disclose
their sponsorships in at least one" post. On aspartame specifically it found at least 35 posts
from 12 health professionals, paid by American Beverage — the lobbying group whose members
include Coca-Cola and PepsiCo — to push back on WHO's announcements. Only 11 of the 35 videos
named the association or "AmeriBev" at the time of posting. Two months later the FTC sent
warning letters over roughly three dozen posts, objecting that disclosures such as a bare #ad
were "inadequate" — an account we take from The Examination's follow-up, not from the FTC's own
documents, which we could not read.
None of this shows anyone was bought: warning letters are warnings, not findings that a law was broken, several dietitians said the sponsor had no editorial control, and Mary Ellen Phipps said the suggestion that "I would compromise my professional ethics for money is insulting". The documented failure is disclosure.
Sugar warnings are not a psychological artefact
WHO's 2015 guideline makes reducing free sugars a strong recommendation — its term for one whose "desirable effects… outweigh the undesirable consequences". Cutting free sugars — those added by a manufacturer, cook or consumer, plus those in honey, syrups and juice, but not the sugars in whole fruit, vegetables or milk — below 10% of energy is also strong, on moderate-quality evidence about dental caries; below 5% is conditional, on very low-quality evidence. The UK's Scientific Advisory Committee on Nutrition, which disagrees with WHO about sweeteners, still restated in April 2025 that free sugars should not exceed 5% of energy. Two committees that read the evidence differently land in the same place on sugar.
Where the claimants have a point is certainty. The 2023 BMJ umbrella review graded 73 meta-analyses covering 83 health outcomes: none reached high-quality evidence. Four — 5% — reached moderate quality: body weight gain from sugary drinks, and ectopic, liver and muscle fat accumulation from added sugars. Eight more reached the tier below, among them type 2 diabetes and coronary heart disease. Most of that evidence is observational and graded accordingly: the cohort, case-control and cross-sectional findings are "graded as 'low' or 'very low' quality owing to the observational study design", the authors write, and heavy sugar intake "may be a marker of an unhealthy diet and lifestyle". They still recommend added sugar below 25 g a day. The certainty is lower than the rhetoric; the direction is not in doubt.
Aspartame at ordinary intakes is not the problem
This part the claim gets right. EFSA's 2013 re-evaluation concluded that aspartame "is safe at current levels of exposure" and kept the acceptable daily intake at 40 mg per kg of body weight as "protective for the general population". (We read EFSA's summary, not the full opinion.)
In July 2023 two WHO bodies reported on the same day; merging them is the commonest error in this topic's coverage. IARC classified aspartame as Group 2B, possibly carcinogenic, on limited evidence for liver cancer in humans and limited animal and mechanistic evidence. JECFA simultaneously reaffirmed the acceptable daily intake of 0–40 mg/kg, finding "no sufficient reason to change" it. IARC asks whether a substance can cause cancer; JECFA asks whether it does at the doses people consume. On WHO's own arithmetic, a 70 kg adult would need more than 9 to 14 cans a day of a diet drink containing 200–300 mg of aspartame to exceed the limit. WHO's Francesco Branca gave the honest middle — "while safety is not a major concern at the doses which are commonly used, potential effects have been described that need to be investigated by more and better studies."
"No health risks" is still false
The cleanest refutation comes from the regulator the claim leans on. EFSA states that for people with phenylketonuria "the ADI is not applicable, as they require strict adherence to a diet low in phenylalanine." That group must avoid aspartame — a small risk confined to one population, but "no health risks" does not survive it.
Beyond that the picture is unsettled, not reassuring. WHO's 2023 guideline "suggests that non-sugar sweeteners not be used as a means of achieving weight control or reducing the risk of noncommunicable diseases" — a conditional recommendation on low-certainty evidence, not a ban, not a safety ruling, and not applicable to people with existing diabetes. SACN recommends "that intake of NSS be minimised" and notes "insufficient evidence to carry out a full risk assessment."
The observational signal is limited. In the French NutriNet-Santé cohort — 102,865 adults, median follow-up 7.7 years — higher aspartame intake was associated with overall cancer at a hazard ratio of 1.15 (95% CI 1.03–1.28) and breast cancer at 1.22 (1.01–1.48); sucralose showed no cancer association. A cardiovascular analysis of the same cohort, which we have only in abstract, linked aspartame to cerebrovascular events at 1.17 (1.03–1.33) and claims only "a potential direct association". These are associations in a volunteer cohort whose participants "were more likely to be women… to have health-conscious behaviours", and whose authors write that "causal links cannot be established by this unique study."
Two committees, one review, different answers
WHO and SACN read the same systematic review and disagreed, because SACN "gives greater weight to RCTs of adequate quality and duration and less weight to PCS evidence… WHO gave greater weight to PCS evidence over RCT evidence."
Weighting the trials, SACN finds that sweeteners used in place of sugars produced a small reduction in body weight in adults — 0.71 kg pooled, low certainty. Cohort studies point the other way, and SACN explains why they might: reverse causality, residual confounding, and the fact that "adjusting for BMI at baseline is unlikely to fully adjust for confounding where NSS use is a marker of dietary concern or restraint." That applies to NutriNet-Santé.
Here is the distinction both sides drop. Restricting the trials to those instructing habitual sugary-drink consumers to swap made the weight effect non-significant, and sweeteners tested against water or placebo showed no effect. Replacing sugar with a sweetener is a different question from adding one, and only the first has ever looked like it works. SACN vindicates nobody: every sweetener used in the UK has passed EFSA safety assessment, and it recommends minimising them anyway.
They are right about food guilt
The "food anxiety" framing distorts a real clinical position rather than inventing one. Several sponsored posts were made in the language of removing stigma — one dietitian on "removing guilt and shame around food", another on letting children have unrestricted access to sweets to "prevent sweets obsession". That position has experimental support: a 2014 study in Appetite, which we have in abstract only, found that short-term restriction of a palatable food increased 37 preschoolers' selection and intake of it, most strongly among children whose parents habitually used restriction — "the use of restriction does not reduce children's consumption of these foods." Thirty-seven children in a laboratory is not a study of adult food guilt, but it is a real result: shame-based nutrition advice is a poor tool.
Why this is a D
Not an E. The sweetener argument's defensible version stands on genuine regulatory findings — EFSA's conclusion, JECFA's reaffirmed limit, the 9-to-14-cans arithmetic — and the "food anxiety" half distorts a position that has evidence behind it. Decisively, "no health risks" is a compression: nobody in the sourced record is quoted saying those words.
Not a C either. The evidence is genuinely split on sweeteners, where two committees read one review and reached different recommendations. It is not split on the other half: reducing free sugars is a strong WHO recommendation that SACN independently reiterates.
D, because both halves take something real and push it one step too far. The uncertainty in the sweetener evidence is genuine — and a lobbying group paid credentialed professionals to sell that uncertainty as reassurance. Being right that the evidence is unsettled is not the same as saying there is nothing to settle.