Who AI cites when patients ask about plastic surgery.

When a patient asks an AI assistant about a facelift or a rhinoplasty, the answer arrives with sources. We wanted to know whose. So we ran the same panel of patient questions through four AI engines on one day, twice, in four metros, and counted which sources each engine cites.

Google’s AI runs on video, social, and Google

Nearly 2 in 5 AI Overviews answers cited Instagram or YouTube (38%; AI Mode is the same at 37%). RealSelf still appears, about 1 answer in 6, but it competes with social feeds, Yelp, and Google’s own properties. The same questions that produce a RealSelf-and-Reddit answer in ChatGPT produce a video-and-reviews answer in Google.

ChatGPT runs on RealSelf and Reddit

47% of ChatGPT’s answers cited RealSelf at least once, more than any other source a practice can act on. Reddit came second at 28%, then the boards and directories behind them: plasticsurgery.org, ABFPRS, Healthgrades. When a patient asks ChatGPT about a procedure or a surgeon, a large share of what comes back is assembled from your RealSelf profile, Reddit threads, and board listings. If your RealSelf profile is stale, that is your first impression.

What kind of video gets cited

The prompts that pulled YouTube citations were not tutorials. They were decision questions: what a facelift costs, how to choose a surgeon at 55, whether anyone has had a bad facelift in Beverly Hills, credential checks. Patient-decision content, not procedure explainers, is what Google’s AI surfaces reach for video to answer. If your videos explain procedures instead of helping people decide, they are aimed at the wrong questions.

Cost questions get answered from practices’ own cost pages

Across the cost and financing bucket, Google’s AI repeatedly cited individual practices’ own cost guides: pages titled “how much is a facelift in Houston” or “rhinoplasty cost in Chicago.” A structured, metro-specific cost page is directly citable material. This is also where the engines split: both Google surfaces cited dozens of individual practice websites; ChatGPT almost never did, preferring the directory layer. Your own site is a citation surface in Google’s AI. In ChatGPT, your representation is whatever RealSelf and Reddit say it is.

Reputation questions get answered from your Business Profile

In Google’s AI Mode, the domain cited more than any other was Google itself, overwhelmingly links into the review viewer showing Google Business Profile reviews. Local reputation questions resolve inside Google, from your Business Profile, without the patient ever leaving. The takeaway is direct: your Google Business Profile, its reviews, their recency and volume, is source material for AI answers about your reputation. Optimize it like the publication it now is.

Different patients, different engines

Platform adoption data (Pew Research) consistently shows ChatGPT usage skewing under 35, while search remains the default across older cohorts. Layered on the citation data: the younger patient researching rhinoplasty or filler is likelier reading an answer built from RealSelf and Reddit; the older facelift consult is likelier reading one built from Instagram, YouTube, and Business Profile reviews. The age-to-engine mapping comes from adoption data, not from these panels.

What this snapshot is and is not

Everything above describes July 23, 2026, and nothing more. One run per panel; engines refresh what they cite, and measuring that movement is the point of running this as a dated series. Gemini is excluded from the findings: 57% of its answers contained no citations at all, too thin to support claims. Prompt sets are frozen and publish as-is, flaws footnoted: one prompt carries a stale metro reference from an earlier draft, and the two panels’ prompts are parallel but not word-for-word mirrors, so we make no metro-versus-metro claims.