When a prospective client asks an AI search platform a question about longevity care, the platform rarely answers from the clinics' own websites. It answers from an intermediation layer: a structured ecosystem of directory sites, comparison content, review aggregators, mainstream press and academic journals that the platforms treat as authority. In most query contexts, the 25 clinics in this study are not the sources AI search reads. They are the subjects AI search describes, using sources written by other people.
A small number of panel clinics, most clearly Fountain Life, have crossed into the intermediation layer themselves through self-published authority content. The other panel clinics appear in that layer as subjects.
AI platforms weight source authority on independence, comparison breadth and external corroboration. A clinic's own website provides none of the three. A page describing a clinic's hormone optimization protocol is a single source with no external comparison. A ranking site that compares dozens of providers against named criteria and updates regularly is, from the platform's perspective, doing the work the platform would otherwise have to do itself. Where the two are in tension, the comparison source wins.
The intermediation layer is tiered. AI platforms route through different tiers depending on what is being asked, and the tiers do not share sources, authority signals or competitive dynamics. A clinic winning citation share in one tier can be entirely absent from another.
The first tier is service-led directories. When a prospective client asks where to obtain a specific procedure or therapy, AI platforms surface specialty operators and category-ranking sites rather than longevity clinics that happen to offer the service. Each service category (NAD+ therapy, peptide therapy, hormone optimization, full-body MRI, stem cell therapy, biological age testing) has produced its own directory ecosystem of ranking sites and clinic-finder databases that exist only to compare providers within that one category. Across the service-led queries in the study, fewer than one in five of the 25 clinics surfaced on any given question. The longevity clinics that offer the service lose category share to the operators and directories that own the category.
The second tier is geo-anchored review aggregators and local listings. When a query names a city or region, the platforms shift toward Google listings, TripAdvisor and local clinic-comparison sites. The pattern is sharpest on the Dubai and Singapore queries, where one platform returned a ranking ordered by Google review count rather than medical credentials, and a physiotherapy practice with a large review count anchored the Dubai longevity list ahead of dedicated longevity clinics. For European luxury retreats, TripAdvisor citations dominate. Local-listing hygiene and review accumulation function as a form of geographic search optimization that influences AI citation directly, independent of any content strategy.
The third tier is comparison content and community discussion. Head-to-head queries pull comparison blogs, specialist review sites and Reddit threads. This is the tier where a clinic's own published comparison content can win citations on queries it is not the subject of. It is also where the dataset's clearest brand-integrity failure occurred: on the Wild Health versus Parsley Health comparison, one platform, unable to resolve "Wild Health" as a distinct medical entity, returned a list of wrong sources including a movie, a dictionary entry and a brand of deodorant. When the comparison tier has thin authoritative sourcing, the platforms fill the citation panel with semantically adjacent noise.
The fourth tier is mainstream press and academic journals, surfaced on evidence-evaluation and skeptical queries. This is the most authoritative layer in the ecosystem, and the layer where no clinic in the study had direct visibility. The question of whether longevity medicine is evidence-based routes to peer-reviewed journals, major research institutions and mainstream press health desks. None of the 25 clinics surface here because the query is not asking about clinics. It is asking about the legitimacy of the field, and AI search answers that question from the strongest available authority.
The testing-infrastructure layer§
A fifth tier became fully visible when the service-category queries were retested with controls for geographic localization. For queries asking where to obtain a single specific test (whole genome sequencing, VO2 max, full-body MRI, biological age), AI platforms route the answer through dedicated testing-infrastructure operators rather than through full-service longevity clinics, even when the clinic offers the test as part of its membership.
The VO2 max query is the clearest case. The platforms returned answers dominated by a fitness-testing franchise network, national locator services and gym chains, with none of the 25 clinics surfacing, despite VO2 max testing being a routine component of any longevity workup. The full-body MRI query routes to scan-specialty operators. The whole-genome sequencing query anchored on a specialty clinic outside the panel rather than on the panel's genomics pedigree leader. A clinic that lists a test as one bullet on a services page loses the corresponding citation to an operator whose entire content footprint is built around that one test and the client's location.
Single-test queries are a structurally different competitive layer from the integrated-care queries where full-service clinics win. The strategic response is to avoid overclaiming against the operators that own the single-test categories.
The non-panel entities AI search treats as category authority§
Three entities outside the study's 25-clinic panel surfaced repeatedly enough, and with enough authority, to illustrate what being cited instead looks like in practice. Each occupies a category position that panel clinics were positioned to compete for and did not win.
TruDiagnostic is a deep-science epigenetics company, not a clinic. It anchors the biological age testing category across three of the four platforms on the strength of a research pedigree no panel clinic offering biological age testing can match: a large body of peer-reviewed clinical-trial validation, academic research collaborators and an exclusive license on a gold-standard pace-of-aging measure. Each peer-reviewed paper functions as an independent third-party citation, which compounds faster than self-published content. A 2026 corporate development sharpened the position further: TruDiagnostic's parent acquired a consumer biological-age brand, and TruDiagnostic's chief executive also leads one of the panel's own clinics, which means the company now surfaces across an interconnected set of adjacent queries that a standalone diagnostics platform could not reach.
TruDiagnostic occupies an unusual structural position. Its platform is the diagnostic infrastructure underlying the 21-year biological age spread finding that anchors the report's evidence position. The same company is winning category citation share and providing the evidence that demonstrates why the category's marketing claims of test precision are unsupported. The operator winning the citation war is publishing the evidence that constrains how the category can responsibly claim accuracy. For panel clinics offering biological age testing as part of their service stack, the citation share they cannot win is held by an operator whose own evidence base argues against the precision claims that would justify clinic-level marketing of the same tests.
Aviv Clinics is a hyperbaric oxygen therapy specialty operator with flagships in Florida and Dubai. On the platform that weights peer-reviewed research most heavily, it anchors the HBOT-for-longevity query as the category leader on the strength of a landmark 2020 study (the first to document reversal of two aging hallmarks in humans under randomized conditions) and an institutional partnership with the world's largest hyperbaric research center. Aviv is largely absent from the same query on the platforms that extract structured content rather than read research framing. A clinic can hold the strongest evidence base in its category and remain invisible on the platforms that do not read research framing.
Parsley Health is a functional medicine telehealth provider that anchors the women's functional medicine query on three of the four platforms, where the panel's own clinics do not. Its authority comes from founder credibility paired with a clear chronic-care editorial positioning the platforms have absorbed and route to consistently. A 2026 business model shift moved Parsley out of the crowded cash-pay concierge tier into a category position no competitor currently matches: it became the first functional medicine telehealth provider to accept major insurance nationwide, at a substantially lower membership price, and the platforms have absorbed the insurance-accepting framing into how they describe it.
Where three of the four platforms anchored Parsley on women's functional medicine queries, the fourth produced a competitive set built entirely from individual practitioner profiles, with no panel clinics surfacing at all. Same query, same brand authority, three platforms agree, one routes to an entirely different intermediation layer. For panel clinics whose category positioning resembles Parsley's, strong cross-platform anchor visibility on one positioning frame does not guarantee anchor visibility on a fourth platform whose editorial preferences route the same query differently.
The common thread across the three non-panel entities is that none won their category positions through marketing. Two won through research, the third through founder credibility and a structural business model change. Each occupies a position a panel clinic could have contested, and the panel clinics did not.
Reddit is the most-cited non-clinic source layer in the dataset after the clinics' own domains. Within the intermediation map, the structural fact is that Reddit is not cited as a monolithic source. The platforms route to specific communities for specific query categories: longevity medicine communities for executive and biological age questions, biohacking communities for protocol-specific service queries, brand-specific communities for cost and value questions, and futurology and skeptical communities for questions about whether the field is legitimate. Clinics should treat Reddit as a structured layer of community authority pages, each anchoring a different query type.
The criticism ecosystem§
The skeptical query layer shapes the first impression a prospective client forms of the longevity industry as a whole. Across the queries about whether longevity clinics are a scam, what their common complaints are, and whether they are worth the money, a coherent criticism ecosystem surfaces: mainstream long-form journalism positioned skeptically toward the category, demographic publications focused on cost, academic critique and review-aggregator complaint summaries. On the complaints query, one platform returned a structured breakdown of recurring grievances: generic non-personalized care, aggressive upselling of unproven treatments, a "cash grab" perception and operational failures.
Any prospective client searching with skeptical framing encounters this ecosystem before they encounter any clinic's own content. The narrative authority on the field's legitimacy has been established outside the clinics, and it is consistently mixed. The criticism ecosystem also cites peer-reviewed academic sources at a higher rate than any other query type, alongside reporting on the 2025 longevity-clinic survey that anchors the report's evidence position. When a clinic gets fewer AI citations than expected, part of the answer is that the most authoritative sources in the ecosystem are positioned skeptically toward the industry, and earning citations from them requires academic credibility.
Where AI search declines to participate§
On Google AI Overviews' suppressed templates there is no intermediation layer because there is no AI-generated answer to cite into. Clients fall back to conventional Google search, where the competitive layer is the standard organic-search ecosystem. For clinics whose clients ask the questions that fall into the suppressed templates, conventional search remains the dominant competitive layer because AI search has declined to participate.
What this means for clinics§
Optimizing a clinic's own service pages will not produce visibility on the corresponding service queries; those queries are won by category directories and specialty operators that have built their entire content footprint around individual service categories. Local-listing hygiene and review accumulation matter, but narrowly: they influence geo-anchored queries and the geographic tier of the intermediation layer, not the category-defining queries that establish a clinic's broader brand position.
Comparison and educational content is the most reachable content investment available to most clinics. It produces citations on queries the clinic is not the subject of, and the supply of credible comparison content in the longevity space is structurally thin enough that a disciplined publishing effort can move citation share within a content-indexing cycle.
The clinics most exposed are those that have invested in service-page optimization without building comparison content, those with thin local-listing presence, and those operating in service categories where category-specific directories have already consolidated authority. For these clinics, the strategic question is which tier of the intermediation layer they intend to win, what authority signal they can bring to it, and whether the editorial discipline required is achievable internally.