Probiotic ads, checked
Mice ran longer. In people the trials are tiny — and a bacteria count is not a symptom.
Short answer: each of these has a real piece of science behind it, and each ad stretches it. The athlete strain made mice run longer, but its one human trial had seven people and found no change. PS128 helped runners’ muscles recover in two small trials — there is no trial for the memory claim. Human-milk oligosaccharides do feed “good” gut bacteria, but that did not change the trials’ main outcomes.
How the evidence flows
Each claim (middle), the studies that tested it (left) — green found it, red found nothing — and the grade they add up to (right). Drawn from the same records as the grades below. Tap to open full size.
Probiotics & gut bioactives (Veillonella, PS128, HMOs) · evidence checked 2026-10-25
What the research shows
Grades are calculated from the studies listed, by fixed rules.
- The probiotic PS128 helps muscles recover after hard exercisePromising
Two small double-blind trials in runners after a half-marathon: strength came back faster and muscle-damage markers were lower. Exercise capacity itself did not change. Runners, not older adults.
- “Olympian gut bacteria” (Veillonella) boost enduranceLab studies only
The ad said “Harvard scientists bottled Olympian gut bacteria (yes, really)”
FitBiomics · Facebook ad, seen Oct 2026
The Harvard finding is real — in mice. In people, the one published trial had seven participants, involved the company’s founders, and found time to exhaustion unchanged. The company also cites a clinical preprint, which has not been peer-reviewed.
- Human milk oligosaccharides (HMOs) improve adult gut healthTested — no effect found
They do feed Bifidobacterium — that rose in older adults. But the trial’s main outcome did not change, and after antibiotics they did not help the gut recover. A count of bacteria is not a symptom.
- Open-label randomized trial, 89 people, 6 weeks (2025) — primary outcome: no difference found. Funded by mixed sources.
- Open-label randomized trial, 89 people, 6 weeks (2025) — secondary outcome: improved. Funded by mixed sources.
- Double-blind randomized trial, 18 people, 8 weeks (2026) — primary outcome: no difference from placebo. Funded by independent sources.
- PS128 supports short-term memory as you ageNo real evidence
The ad said “Neuralli Cognition+ may help support short-term memory and muscle strength with clinically studied psychobiotic strains…”
Bened Life · Facebook ad, seen Oct 2026
We found no trial of PS128 for memory in older adults (PubMed, 2026-10-25). Its human trials are in runners, and in pilot studies of neurological conditions.
- Human-milk bioactives work better than bovine colostrum because the body sees them as “self”No real evidence
The ad said “Unlike bovine colostrum, our bioactives were originally discovered in human milk, which means they are seen as “self” by your body.”
Kēpos · Facebook ad, seen Oct 2026
We found no trial comparing them. For what colostrum does have evidence for, see /colostrum.
Before you buy
- A probiotic’s evidence belongs to its exact strain, at the dose tested — check the strain code on the label against the trial.
- “Clinically validated” should point to a published, peer-reviewed trial in people like you, not a preprint or a mouse study.
- If you are immunocompromised, ask your doctor before taking live bacteria.
Your next step
- If you try itFor muscle recovery after hard exercise: the PS128 strain, as testedTwo small double-blind trials in runners found faster recovery after a half-marathon. That is the use with evidence — not memory.
- What has evidence insteadFor gut health: fibre from food firstThe HMO trials raised good bacteria counts without changing their main outcomes. Plants, beans and whole grains feed the same bacteria.