What the agent can and cannot do
The same capability map across all six markets. The columns are identical, because intended purpose is the universal trigger.
| What the agent does | AU | EU | UK | US | IN | AE |
|---|---|---|---|---|---|---|
| Safe to run on the patient path | ||||||
| Neutral fertility / IVF education, not patient-specific | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Clinic info, pricing, insurance and coverage Q&A | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Capture intake details and non-clinical lead qualification | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Book, reschedule, remind, follow up, re-engage | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Route by what the patient says they want | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Grey zone, depends on framing | ||||||
| Answer a symptom question in general educational terms | ⚠ | ⚠ | ⚠ | ⚠ | ⚠ | ⚠ |
| Interactive symptom-checker or quiz that returns a personal read | ✕ | ✕ | ✕ | ✕ | ✕ | ✕ |
| Crosses the line, becomes a regulated device | ||||||
| Interpret this patient's symptoms or assess their case | ✕ | ✕ | ✕ | ✕ | ✕ | ✕ |
| Assign clinical urgency or priority ("triage") | ✕ | ✕ | ✕ | ✕ | ✕ | ✕ |
| Suggest a diagnosis, or tell a patient no doctor is needed | ✕ | ✕ | ✕ | ✕ | ✕ | ✕ |
General education (what is AMH, what does an IVF cycle involve) is fine. The moment a page or agent takes a patient's inputs (my AMH is 0.8, I am 39, trying for 14 months) and returns an interpretation of their situation, it has acted on individual data toward a clinical conclusion and becomes a device. Keep content one-to-many. General in, general out.
By jurisdiction
Australia
TGA
The strictest line of the six, so it effectively sets the bar for the whole product. Regulation is triggered by the manufacturer's intended purpose, assessed objectively from how the product is designed, described, marketed and used. The February 2026 AI guidance explicitly names chatbots that use large language models to apply clinical triage protocols as in-scope. The clinical decision support carve-out is narrow and oriented to clinician-facing tools, not patient-facing symptom interpretation.
Education, intake, qualification, booking, follow-up, and routing by stated interest. The agent never reasons clinically on the patient path.
Any symptom interpretation or urgency call. Full sponsor and manufacturer obligations, ARTG inclusion, likely Class IIa or higher.
Treat the amber row as red-leaning here. Do not let the agent free-reason on symptoms at all. Lock per-agent Intended Purpose statements before any Australian deployment advances.
European Union
MDR + AI Act
Structurally the worst place to be misclassified. Under MDR, any software with a medical purpose is a device, and Rule 11 is severity-based, so almost no standalone medical-purpose software stays Class I. The AI Act stacks on top: medical-device AI is high-risk, and a compliant product must satisfy both MDR and the AI Act. The Digital Omnibus package agreed in late 2025 pushes the AI Act compliance dates back: standalone high-risk AI to 2 December 2027, and AI embedded in MDR or IVDR devices to 2 August 2028. Rule-based software with no learning component is not covered by the AI Act at all.
The neutral knowledge layer relies on the intended purpose too generic line. Education that does not act on an individual's data is not a device.
A symptom-interpreting agent is both a device (likely IIa+) and a high-risk AI system, triggering dual conformity assessment.
Keep public-facing decision logic deterministic and rules-based. The funnel never needed clinical reasoning, so the moat and the regulatory firewall are the same wall.
United Kingdom
MHRA
Post-Brexit Great Britain runs its own regime but keeps the same intended-purpose logic and is aligning with EU MDR and IMDRF. The MHRA's Software and AI as a Medical Device Change Programme is one of the most active in the world. Post-market surveillance regulations are already in force. The draft Medical Devices (Amendment) Regulations 2026, published 11 May 2026, introduce risk-proportionate classification, Predetermined Change Control Plans for SaMD and AI, an International Reliance Pathway, and Regulation 19D prohibiting misleading or unsubstantiated promotional claims. A dedicated AI framework is expected during 2026.
Same green zone as the EU and Australia: non-clinical education, intake, qualification, routing.
Symptom interpretation classifies as a device. Reg 19D adds separate exposure on co-branded efficacy claims.
The International Reliance Pathway means a future EU CE or US clearance would ease entry. Watch Reg 19D for any co-branded clinic content with implied efficacy.
United States
FDA
The most permissive regime for clinician tools, and it just got more so, but that does not help a patient-facing use case. The January 2026 CDS final guidance, which supersedes the 2022 version, broadens the non-device clinical decision support exemption: it removed the automatic time-critical equals device rule and now allows single-recommendation tools under enforcement discretion. But the guidance remains focused exclusively on tools that support a healthcare professional. It is silent on consumer-facing chatbots, symptom checkers, and patient decision support, and the earlier enforcement-discretion proposals for consumer tools were never reinstated.
Education, scheduling, qualification. The same green zone as everywhere.
Patient-facing symptom triage does not fit the CDS exemption, which is built for clinician-facing software. Likely device, unsettled.
The permissive headlines are about clinician copilots, not patient chatbots. For a patient-facing product the line is functionally the same as Australia.
India
CDSCO
Draft but directionally clear. The October 2025 draft guidance confirms that software performing a medical purpose falls within the MDR 2017 regime, classified A to D on intended use, information significance, and situation severity. Crucially, the guidance states that administrative and non-clinical software, such as billing or appointment management systems, stays outside scope. A triage-capable agent would likely be Class C under central CDSCO licensing. The Telemedicine Practice Guidelines 2020 separately bar platforms from making clinical decisions, so both regimes point the same way.
The data management and general communication carve-out covers the product. Appointment, intake, education, routing all sit here.
Symptom interpretation likely Class C, central licensing, AI/ML validation burden.
The schedule or connect to a human, never decide output design is exactly what keeps the system in the carve-out. Hold that boundary throughout the build.
United Arab Emirates
EDE
Reliance-based. Federal Decree-Law 38 of 2024 took effect 2 January 2025 and consolidated regulation under the Emirates Drug Establishment, which by early 2026 absorbed the medical-device functions previously held by MOHAP. The UAE uses GHTF-style classification and relies primarily on approvals from reference regulators such as EU CE marking or US FDA clearance. The Abu Dhabi DOH and Dubai DHA each publish AI healthcare policies that sit inside the broader federal and emirate framework, including data protection.
Non-clinical education, intake, routing. Same green zone.
Reliance imports the EU and FDA strict line by reference, so there is no softer back door for triage-capable software.
A prior EU or FDA status would smooth entry if registration were ever needed, but the non-clinical design avoids that. The binding constraint in the UAE is data residency, not device classification. See Hosting.
Others worth flagging
Singapore (HSA). SaMD regulated by the Health Sciences Authority on the same IMDRF logic, with reliance routes and active digital-health guidance. A common medical-tourism and expansion node. Non-clinical funnel stays out.Saudi Arabia (SFDA). Medical Device Interim Regulation under the Saudi FDA, GHTF-aligned, paired with the Saudi PDPL. The natural Gulf parallel to the UAE, same reliance posture.Canada (Health Canada). SaMD under the Medical Devices Regulations with an IMDRF risk model and dedicated machine-learning guidance. Same green and red zones.The takeaway across all of them is identical: the non-clinical, route-by-stated-interest design clears every IMDRF-aligned market without registration. New markets change the data and advertising work, not the device line.
