Driver and partner onboarding checks for gig platforms
Ride-hailing, delivery and service platforms onboard partners from a phone: a selfie, an ID, a driving licence and vehicle papers. TrueSecurix checks the person with identity checks and the papers with image checks, and returns pass, review or flag with the evidence behind each answer.
In our held-out test of 31,966 images, 0 of 2,526 genuine business photos (shops, markets, receipts, documents) were flagged and 87.8% of manipulated images were caught. Read how we measured
What onboarding and safety teams at platforms are seeing now.
What partner onboarding fraud looks like today
Someone else behind the account
A partner onboards with a borrowed or face-swapped selfie, so the person doing the work is not the person who was checked.
A person who does not exist
A fully generated face paired with a made-up Aadhaar or PAN number.
Edited licences and vehicle papers
A name, a date or a vehicle number changed on a photo of a driving licence or a registration certificate.
Screens and recordings at the camera
A photo of a photo, a screen held up to the camera, or a recorded video fed in through a virtual camera.
Image and identity checks run on the same account and the same credits: one credit per check.
Which checks apply, and what you get back
Each image goes to POST /v1/image/authenticity; selfies, IDs and statements go to POST /v1/verify. Every response carries an outcome and the evidence behind it, and every check is written to your history.
| What you receive | Check | What the response gives |
|---|---|---|
| Partner selfie | Deepfake selfie detection | Whether the face was swapped or generated, and a passive signal for a photo or screen held to the camera. |
| Selfie and ID photo | Identity checks | Whether the selfie matches the photo on the submitted ID. |
| Aadhaar and PAN | Checksum validation | The field and the reason when a number fails its own check digits or format. |
| Driving licence and vehicle registration, as photos | Image authenticity | Real, AI generated, AI edited or digitally edited, with the indicators and the located edited areas. |
| Device signals at capture | Injection checks | Virtual cameras, emulators and screen feeds, read by our capture SDK. |
Nobody is rejected on a guess. The model tells you which files a person should look at.
Three outcomes your onboarding team can act on
The decision field turns the answer into an action. Route on it directly in your onboarding system.
Accept it
No sign of a deepfake, a failed checksum or an edited document. The partner continues through your normal onboarding.
A person decides
Some evidence points to a problem, not enough to accuse a genuine partner. An agent takes a short look, guided by the evidence.
Treat as manipulated
Strong evidence: a swapped or generated face, a number that fails its own checksum, or an edited document with the changed areas marked.
How a team rolls it out
Most of the work is wiring three outcomes into the onboarding system you already run. You can build all of it before you spend a credit.
One request, one image, one answer
curl https://truesecurix.com/v1/image/authenticity \ -H "X-API-Key: $TRUESECURIX_API_KEY" \ -F "image=@driving_licence.jpg" \ -F "reference=partner-50624"
Field by field in the API reference.
- Build against a test key. A test key returns a simulated answer at no charge, including the exact review shape when you send
reference=review, so your onboarding system handles all three outcomes before any live image is sent. - Run a proof of concept on your own labelled images. A folder of real and manipulated selfies, IDs, licences and vehicle papers from your own files, with a CSV of labels, goes through the batch script. You get accuracy per class, the confusion matrix, genuine images wrongly flagged and the share sent to review, on your data.
- Send results into your onboarding system. Signed webhooks post
check.flagged,check.reviewandcheck.completedto your HTTPS endpoint, signed with HMAC-SHA256 and retried for about a day. - Route reviews to a person. A review carries the class the evidence leans toward and the indicators behind it, so a reviewer knows what to look for. Every check stays in your history with its model version, and exports to CSV.
Measured on images it never saw
The image model was measured on business photos of shops, markets, receipts and documents, plus general photos, in a held-out test of 31,966 images never used in training, threshold setting or model choice.
Your own images are the measure that counts for your use. A proof of concept on your own labelled images gives you the same figures on your data, with the per-image CSV behind every number.
Questions risk, operations and engineering teams ask first.
Questions, answered
Do you check licence and vehicle numbers with the government?
No. We check the licence and vehicle paper images for generation and editing, and validate Aadhaar, PAN and passport numbers by their own check digits and format. We do not claim any government database lookup or authorisation that we do not hold.
Will genuine partners be rejected?
On our held-out set, no genuine selfie was rejected and about 12% were sent to review. A review goes to a person in your team, never to an automatic rejection.
Does it replace our liveness or video KYC provider?
No. The deepfake and injection checks run alongside your existing liveness or video KYC flow and add a layer for face swaps and generated faces.
How accurate is it on our partners' documents?
The image model was measured on business photos of shops, markets, receipts and documents, plus general photos, in a held-out test of 31,966 images. It has not been measured on licences or vehicle papers specifically. A proof of concept on your own labelled images is how you measure it for your use.
Check partners before their first job
Try a photo in the live demo now, or create an account for 100 free checks. No card is needed, and test keys are free while you build.
Running lakhs of checks a month? See enterprise terms and a proof of concept.