Deepfake detection for healthtech
Catch face swapped patients and practitioners on telehealth calls, forged prescriptions and medical records, and cloned clinician voices. Raid AI checks video, documents, images and voice with zero retention available.
4
Video platforms covered for consultations
4
Video platforms covered for consultations
<50ms
Audio detection latency
99.7%
Audio detection accuracy
Remote care moved identity onto a screen. A patient is a face on a video call, a prescription is a PDF, and a clinician is a voice on the phone to a pharmacy. Each of those can now be generated. Raid AI checks the consultation, the documents and the voice at the moments a healthtech platform decides who it is treating and what it is authorising, and it can run without keeping any patient media.
5 attacks being run against healthtech today
These are not projections. Each one has already been used against organisations in this position.
| Attack | Method | Modality | Who it lands on |
|---|---|---|---|
| 1Patient identity fraud in telehealth | A face swap on a video consultation lets someone be treated, prescribed or certified in another patient's name. |
| Who it lands on: Telehealth providers |
| 2Forged prescriptions and medical records | Prescriptions, sick notes and medical reports are edited or generated whole to obtain medication, leave or a payout. |
| Who it lands on: Pharmacies, insurers, employers |
| 3Cloned clinician voices | A cloned voice of a known doctor calls a pharmacy or a lab to authorise a prescription or have results released. |
| Who it lands on: Pharmacies, labs |
| 4Generated clinical images | Wound photos and scan images are generated or edited for remote assessments and health claims. |
| Who it lands on: Remote care and claims teams |
| 5Fake practitioners on the platform | A forged licence, a generated ID photo and a face swapped verification call put an unqualified person in front of patients. |
| Who it lands on: Credentialing teams |
How Raid AI answers it
Telehealth session checks
Raid AI joins the consultation as a participant and scores the video and voice while the call runs, so a swapped face is caught before a prescription is issued.
Prescription and record forensics
Prescriptions, medical reports and invoices are checked for edited fields, pasted text and generated pages before they are dispensed or paid.
Voice checks on authorisation calls
Calls asking a pharmacy or a lab to authorise or release something are checked for a cloned voice while the caller is still on the line.
Practitioner onboarding
The selfie video and documents a new clinician submits are checked for face swaps, injected video and forged licences before they see a patient.
- Patient data stays where you put it
- Joins consultations held on Microsoft Teams, Zoom, Google Meet and Webex and flags a face swap or a cloned voice during the call
- Document forensics on prescriptions, reports and invoices, with the edited region marked
- Zero retention available, so patient media is analysed and discarded once the verdict is returned
- On premise deployment for providers whose patient data cannot leave their own infrastructure
What this looks like on a normal day
Composite situations drawn from the pattern of reported attacks, not accounts of named customers.
Illustrative scenario
The patient who was someone else
A returning patient joins a video consultation
They ask for a repeat prescription of a controlled drug
Raid AI flags the face as swappedFlagged
The clinician verifies identity before prescribing
Illustrative scenario
The doctor who called the pharmacy
A pharmacy gets a call from a doctor it knows
The caller authorises an urgent prescription
Raid AI flags a cloned voiceFlagged
The pharmacist calls the practice back on its listed number
Built for environments where security is not negotiable
From air gapped networks to strict data residency rules.
Retention you control
You choose how long media is kept. With zero retention, media is analysed and discarded once the verdict is returned and nothing is written to Raid AI storage. Otherwise media is kept for a window you set, 30 days by default, and then deleted. None of it is used to train a model, and enterprise tenants can keep their records inside their own database or infrastructure instead. Media goes only to the detection engines your tenant is configured to use, and only to produce the verdict.
Runs in your infrastructure
For air gapped and sovereign environments, the full detection stack runs inside your own network with no external connectivity.
TLS 1.3 encryption
All communication with the cloud API is encrypted using TLS 1.3. Media is protected from the moment it leaves your network.
Scoped API keys
Every integration is gated by a scoped key with configurable rate limits and IP allowlisting. Keys can be rotated at any time.
SOC 2 principles
Built around the five SOC 2 trust service criteria: security, availability, processing integrity, confidentiality and privacy.
Regional deployment
Cloud customers choose the region where their requests are processed. Media never leaves a specified jurisdiction.
Defence in depth
Infrastructure, application and model layers are each protected independently. A compromise at one layer does not expose customer data.
Audit logging
Every API call is written to a tamper evident audit trail, so you can reconstruct exactly what was processed, when, and by which key.
