Body scanning in spas is moving from speculative wellness theatre towards a real operating question. Midjourney, best known for image generation, has announced a warm-water body-composition scanner and a San Francisco spa planned around it. The concept is visually arresting: bathing, recovery and repeated measurement in one journey. Yet the professional test is not whether the idea looks futuristic. It is whether guests can understand what is measured, choose freely, receive appropriately limited information and leave without losing control of deeply personal data. For spa leaders, “general wellness” should mark a boundary, not provide a blank cheque for medical language, opaque personalisation or unlimited data use.
What has been announced—and what remains planned
Spa Business reported on 21 July that TLee Spas and Wellness had revealed a design concept for Midjourney involving spa, fitness and body mapping. That report is a news signal, not proof that a public venue, production scanner or health outcome already exists.
Midjourney’s own launch announcement describes a future San Francisco spa with hot tubs, saunas, cold plunges and warm-water scanning. A more detailed development update dated 23 June 2026 places the spa in Union Square and lists November 2027 as the target grand opening. It says the current scan takes more than 20 minutes, while a one-minute scan is a future goal. The same page labels all dates as targets and future technical capabilities as goals rather than current features.
That distinction is fundamental. Published images are concept renders. The venue is in design, the device is in development and commercial operation has not begun. No independent public evidence reviewed for this article establishes measurement performance, repeatability across guest populations, guest acceptance, operational throughput or health benefit. Midjourney’s broad public-health ambitions should therefore remain ambitions, not promises repeated by spa marketers.
General wellness is a scope boundary, not an approval claim
Midjourney says the first product is intended to provide body-composition information, not diagnose, treat, cure or prevent disease. Its development page reports that a June 2026 FDA response classified the proposed scanner as a Class II body-composition analyser exempt from 510(k) premarket notification within its stated use. The company itself says this is not a stamp of approval. The underlying agency correspondence was not independently reviewed for this article, so the classification details are attributed to Midjourney rather than presented as a separate Spa Awards finding.
The US Food and Drug Administration’s January 2026 general-wellness guidance explains that certain low-risk products and software can encourage a healthy lifestyle when they are unrelated to diagnosis, cure, mitigation, prevention or treatment of disease. It does not make every wellness product unregulated, validate every measurement or substantiate every outcome. Intended use, claims, function and risk still matter.
The Federal Trade Commission adds a separate marketing boundary. Its Health Products Compliance Guidance says objective health-related claims need appropriate substantiation and that marketers must consider the overall impression created by words, visuals and context. A technical classification or disclaimer does not repair an otherwise misleading claim. In a spa, the room design, staff script, app labels, before-and-after graphics and package name can imply diagnosis or improvement even when the small print says “general wellness”.
Keep three propositions separate
- Measurement: what the device directly estimates, in what units, with what uncertainty and under what conditions.
- Interpretation: what a qualified person or validated rule may responsibly explain about that result.
- Outcome: whether any recommended action reliably changes health, wellbeing or behaviour.
Evidence for one proposition does not automatically prove the next. A detailed image is not necessarily an accurate estimate; an accurate estimate is not automatically meaningful for one guest; and information that is meaningful does not by itself improve an outcome.
Body scanning in spas starts with a precise purpose
Before selecting hardware, operators should write one plain-language purpose statement. For example: “The service estimates selected body-composition measures so an adult guest can view change over time for general wellness.” Every field, screen, staff action and marketing sentence should be tested against it.
If the journey starts making disease-risk inferences, directing treatment, triaging symptoms or implying that a scan can find illness, the purpose has changed. If data collected for a guest’s report is later used for research, product training, advertising or population comparison, the purpose has changed again. New purposes need their own authority, evidence, explanation and choice; they should not be hidden inside one broad acceptance screen.
A useful pre-launch register lists every output and prohibited interpretation. It names who may see raw images, derived measurements and trend summaries; which staff may explain them; what referral route exists when a guest asks a medical question; and which phrases are never used. “Within the intended general-wellness scope” is clearer than “medically accurate”. “This estimates body composition” is different from “this detects a problem”.
Consent must be a service step, not a buried form
A guest may arrive expecting bathing or recovery and encounter scanning as part of the architecture. That makes timing and choice especially important. Consent should come before changing, showering or entering a device room. The guest should understand the physical experience, what is measured, which outputs appear, who receives them, how long they remain available and how to decline.
Choice is not meaningful if refusing the scan cancels an unrelated spa visit, creates embarrassment or removes a benefit that does not depend on data. Operators should design a dignified non-scan route, train staff to present it neutrally and measure whether people feel pressured. The Journal’s review of responsible retreat principles offers a relevant precedent: qualifications, safety, privacy and transparent booking need operational evidence, not reassuring copy alone.
Separate the permissions
One choice can authorise the immediate scan. Separate choices should cover retaining results, comparing future visits, sharing with another professional, using de-identified information for product improvement and using any image in research or publicity. Withdrawal and deletion routes should be as visible as enrolment. A staff member should be able to explain what deletion removes, what may have to be retained and what has already been transformed into a genuinely non-identifiable form.
Map the data before designing the personalised journey
Body-composition images and inferences can be more sensitive than a standard booking record. The California Privacy Protection Agency’s consumer FAQ lists health information and biometric information used for identification among examples of sensitive personal information. Whether a particular spa dataset meets a legal definition depends on what it contains, how it is used, where the service operates and which rules apply. International operators need market-specific advice; “not diagnostic” does not mean “not personal”.
Midjourney says its initial direct-to-consumer model will generally sit outside HIPAA’s strict scope, while it intends to build strong safeguards and give users access and deletion controls. That is the company’s stated plan, not an audited implementation. More broadly, being outside one health-privacy law does not create a privacy vacuum. Consumer, data-protection, security, contract and sector duties may still apply.
The stable NIST Privacy Framework 1.0 is a voluntary risk-management tool rather than law, but its logic is useful: identify the data and its context, govern responsibility, control processing, communicate with people and protect the information. A spa implementation should map the full lifecycle: booking identifier, eligibility answers, raw sensor data, reconstructed image, derived measurements, staff notes, app account, vendor logs, backups, analytics exports and deletion records.
For each element, record the purpose, system, owner, access group, location, transfer, retention period and deletion method. Vendor contracts should address incident reporting, subcontractors, model training, secondary use, export, return and verifiable deletion. “Encrypted” is one control, not a complete privacy programme.
Measurement quality comes before personalisation
A premium interface can make uncertain information appear definitive. Before any guest-facing launch, operators should request evidence for repeatability, comparison against an appropriate reference method, tested populations, exclusions, environmental conditions, calibration, software versions and error handling. They should know whether a change between visits exceeds expected measurement variation or merely looks different on a vivid image.
Guest-facing reports need units, date, device version, relevant conditions and understandable uncertainty. Trend views should not turn small fluctuations into red warnings or celebratory green scores without validated thresholds. Staff need a script for “we do not know”, an escalation route for unexpected results and a firm boundary against diagnosing from an image.
Personalisation should begin with low-stakes, reversible choices. Adjusting room temperature, explanation style or whether results are displayed may require less evidence than recommending a programme because of a body map. The more a recommendation affects spending, self-image, diet, exercise or healthcare decisions, the stronger the evidence and professional oversight should be.
The physical spa operation still matters
Data governance cannot replace wet-area safety. Midjourney identifies slips, skin irritation, discomfort, water conditions, supervised exit and operator presence among development considerations. Operators evaluating any similar service should integrate device risk with pool-water management, infection control, cleaning chemistry, electrical safety, emergency access, privacy while changing, accessibility, staff training, maintenance and downtime.
Throughput claims should include the entire cycle: explanation, consent, changing, showering, scanning, exit, cleaning, reset, report generation and questions. A nominal scan time is not room capacity. The operational framework in the Journal’s analysis of wellness beyond the spa suite is relevant here: design intent becomes credible only when commissioning, maintenance, access and evidence ownership are visible.
What does this mean for spa and wellness professionals?
For owners and investors, approve a purpose and evidence plan before approving the spectacle. Separate the economics of the spa, scanner and data platform. Model consent rates, non-scan journeys, full room-cycle time, maintenance, qualified staffing, insurance, security and incident response—not only theoretical scan volume.
For spa directors, design hospitality around freedom of choice. Test the journey with people who are anxious about enclosed spaces, body image, water, technology or data use. Make the exit visible, the non-scan option normal and the escalation route practical.
For therapists and guest teams, define scope sentence by sentence. Staff should know what the system measures, what it does not establish and when to refer a guest to an appropriately qualified independent professional. Sales targets must not reward exaggerated interpretation.
For privacy and technology leaders, inventory raw and derived data, minimise collection, separate permissions, restrict access, test deletion and rehearse an incident. Assess each vendor and transfer rather than assuming one platform policy covers the whole journey.
For marketers, review the net impression. Images of anatomical detail, clinical language, uniforms, expert endorsements and future health ambitions can create implied claims. Label concept visuals, planned dates and unvalidated features clearly.
For awards researchers and editors, ask for operating evidence rather than treating innovation as merit by itself. A concept reveal, directory listing, regulatory category or research protocol does not prove public launch, guest benefit, nomination, judging or an award.
A 30-day pre-launch review
Week one: purpose and claims. Write the intended use, prohibited uses and every express or implied claim across web pages, scripts, visuals and app screens. Classify each statement as verified now, planned, conditional or unsupported. Remove claims that outrun evidence.
Week two: guest journey and physical risk. Walk both scan and non-scan routes. Time the full cycle. Test consent, changing privacy, accessibility, supervised exit, cleaning, water controls, device failure and escalation. Record owners and stop conditions.
Week three: data lifecycle. Map every field and derived output from collection to deletion. Confirm access, retention, backups, vendors, cross-border transfers, research use and incident notification. Run access, export and deletion requests with test accounts.
Week four: measurement and staff readiness. Review validation evidence with appropriately qualified technical and regulatory advisers. Test report comprehension and uncertainty language. Observe staff conversations, correct scope drift and require final approval from operations, safety, privacy and claims owners.
What remains unproven
The reviewed public material does not establish that the proposed Midjourney spa will open on schedule, that a production scanner will match concept descriptions, or that commercial guests will accept repeated scans. It does not publicly establish accuracy across populations, useful thresholds for individual change, improved behaviour, reduced disease, lower healthcare costs or better spa outcomes. FDA guidance explains a regulatory boundary; it is not proof of efficacy. An IRB-reviewed image-collection study protects research participants within that protocol; it is not evidence that a future commercial service works.
The concept remains valuable because it exposes a choice facing the industry. Spas can make intimate measurement feel welcoming, but warmth and beauty increase the responsibility to be clear. The strongest innovation would not be an invisible scan. It would be a visible trust architecture: limited purpose, meaningful choice, measured claims, competent interpretation, secure data and a service that remains hospitable when a guest says no.
Frequently asked questions
Does “general wellness” mean a body scanner is FDA approved?
No. General-wellness guidance describes a scope and enforcement framework. Midjourney says its planned body-composition analyser is Class II and 510(k)-exempt within its intended use, while also stating that this is not a stamp of approval.
Can a spa call body-composition output diagnostic?
Not on the evidence reviewed here. Midjourney states that its initial intended use excludes diagnosis and treatment. Any operator should obtain product-specific regulatory advice and keep staff and marketing inside the verified purpose.
Is one consent form enough?
One document may record choices, but scanning, retention, comparison, sharing, research and publicity are different purposes. Clear separate choices make the guest’s control easier to understand and audit.
Does HIPAA determine every privacy obligation?
No. HIPAA covers defined entities and situations in the US. Consumer and data-protection rules, contractual duties and other market-specific requirements may still apply, especially when information reveals or supports inferences about health.
Do directory links indicate use of body scanning or an award?
No. Readers can browse Canyon Ranch and Sensei Porcupine Creek for wider US destination-wellness discovery. Directory presence does not establish a scanner, partnership, site visit, nomination, judging or award result.
