Defy Medical leads AI-generated answers about peptide clinics and weight-loss providers with a 12.4% appearance rate, but that lead does not hold up statistically, according to 5WPR's new Peptides AI Visibility Index 2026. Ro, Hims, BPC-157 and a non-brand entity called Overview all sit inside the same confidence interval, meaning four AI engines cannot currently agree on a single category leader.
The study asked ChatGPT, Gemini, Perplexity and Claude 48 questions people actually type before choosing a peptide clinic, a GLP-1 provider or a compounding pharmacy, three times each, producing 558 measurable answers. The finding matters for any brand that assumes strong Google rankings still control what a buyer sees: in this category, the AI answer a consumer reads is often built from different sources than the page Google shows.
What Does the AI Visibility Leaderboard Show?
Defy Medical, Ro, Hims, BPC-157 and Overview all landed within a few points of each other, each named in roughly 10% to 12% of the 558 answers returned. Their 95% confidence intervals overlap, so the study cannot say any one of them has a real lead over the others. Below that top tier, Peptide Sciences Corp, Novo Nordisk and the FDA itself all appeared in 7.7% to 8.6% of answers, mixing clinics, drugmakers and a federal regulator into a single result set. That mix is itself a finding: a consumer asking an AI engine "which peptide clinic is best" is as likely to be handed a drugmaker's name or a regulatory citation as an actual clinic.
Why Do the Four AI Engines Disagree So Much?
ChatGPT named Defy Medical in 26 of 128 answers, while Gemini named the non-brand entity Overview in 56 of 144 answers and the other three engines never mentioned it once. Claude, built by Anthropic, returned the most conservative pattern of the four, naming several highly-ranked category brands zero times across 142 answers. A brand chasing one blended "AI visibility score" would miss this entirely: the same brand can look strong on one engine and invisible on another, because each engine uses its own web-search tool, its own training data and its own answer format. 5WPR saw a similar engine-by-engine split in its Med Spa & Aesthetic Medicine AI Visibility Index 2026, where provider visibility also varied sharply by engine rather than moving together.
Why Are Brands Named but Rarely Cited?
Being named in an AI answer is not the same as being cited as its source. Among the 555 answers where the engines showed retrieval evidence, the FDA's own site, fda.gov, was cited in 18.9% of them, the highest rate of any domain measured. The leading commercial brand, Fountain Life, reached only a 5.9% owned-site citation rate, and several visible brands, including Zepbound, Saxenda and Victoza, were cited from their own domains in 0% of sourced answers. Meanwhile a third-party publisher, formblends.com, was cited in 32.8% of sourced answers, a higher rate than any brand in the category carried on its own site.
What Should Brand and Reputation Teams Do Next?
A brand can appear often in AI answers and still have almost no control over what those answers say, because the engines are pulling from third-party sites rather than the brand's own pages. That gap is a reputation exposure as much as a marketing one: if a third-party publisher or a forum thread is doing the explaining, the brand did not write the explanation a consumer is reading. 5WPR's supplements and health brand PR practice recommends three starting moves for any brand named in this study: publish pages that directly answer the 48 measured question types, in language the brand controls; check appearance and citation rate engine by engine rather than as one blended number; and track first-mention position, since a brand named late in an answer is read by far fewer people than one named first.
CONCLUSION
The peptide and weight-loss category shows what AI search is doing to brand visibility across health and wellness broadly: appearance is fragmented across engines, and citation is concentrated on sites brands do not own.
