Video enhancement API for insurance claims

The privacy exposure hiding in claims footage

A single claim video can contain identifiable faces, license plates, home interiors, and sometimes injuries. When you send that to a third-party tool, two common vendor practices become a liability.

First, AI-training clauses in the terms of service — many consumer-grade tools may use your uploads to train their models unless you opt out of a buried clause. Second, retention that is vague or indefinite, leaving claimant footage on a vendor's servers with no clear end date.

On February 23, 2026, 61 data protection authorities published a joint statement on AI-generated imagery, flagging concern about AI systems that can generate or manipulate realistic video of identifiable people without consent, and calling for transparent data handling and accessible deletion. Insurers already carry data-protection duties to claimants; that statement raises the scrutiny any enhancement vendor should expect.

Claims footage also rarely stays in one place — it can move to special-investigations units, defence counsel, and reinsurers. Every hop multiplies the exposure of whatever the original vendor did with it. Starting from a vendor that never trains on the footage and deletes it on a fixed, provable schedule keeps that chain clean from the first upload.

What insurers and insurtech builders need

• An API, not an app. Enhancement has to run inside your claims platform or adjuster app, automatically on upload — not in a desktop tool someone opens by hand.
• Predictable per-claim cost. Per-second pricing and credits that never expire keep unit economics clean, even when claim volume is seasonal or spiky.
• Retention control and proof of deletion. A 30-day auto-delete you can shorten or trigger on demand, with a signed deletion certificate and audit log on the Secure tier.
• A DPA on request and a clear no-training, no-resale commitment for your vendor review.

Clearer footage, faster decisions

Behind every claim video is a decision: is this damage consistent with the reported incident, what does that licence plate read, who is actually in frame? Those calls turn on detail that low-light phone footage and compressed dashcam clips tend to bury. Upscaling to 1080p or 4K, sharpening, and restoring faces and edges surface what is genuinely in the frame — which helps adjusters approve legitimate claims faster and gives special-investigations teams a clearer look when something doesn't add up.

Because it runs as an API, that clarity is automatic. An adjuster opens an already-enhanced clip inside your claims platform instead of squinting at a blurry original or bouncing it through a desktop tool. Enhancement becomes part of intake, not a manual step someone has to remember — and it scales with claim volume rather than with headcount.

How BetterVideo compares

Topaz Video AI is a strong desktop app, but it can't sit inside an automated claims pipeline. API-first competitors such as WaveSpeedAI expose pay-per-use cloud enhancement, but make no headline privacy or retention guarantee in their own marketing — which is exactly the gap that matters for claimant data. Consumer tools (VanceAI, Media.io, HitPaw) carry the training-clause and vague-retention issues above.

BetterVideo pairs API-first with privacy-first: one call enhances the footage on GPU, and it is never trained on, never sold, and auto-deleted — with signed deletion certificates on the Secure tier. See pricing, the API docs, our data-privacy approach, or the legal & eDiscovery API. For committed volume and a DPA, talk to sales.

One API call — curl, Python & Node

Submit a video by URL, then poll GET /v1/jobs/{id} until done (or receive a signed webhook), and fetch a time-limited download link:

curl -X POST https://api.bettervideo.io/v1/jobs \
  -H "Authorization: Bearer YOUR_KEY" -H "Content-Type: application/json" \
  -d '{"video_url":"https://.../clip.mp4","resolution":"1080p"}'

import requests
r = requests.post("https://api.bettervideo.io/v1/jobs",
    headers={"Authorization": "Bearer YOUR_KEY"},
    json={"video_url": "https://.../clip.mp4", "resolution": "1080p"})
job = r.json()  # poll GET /v1/jobs/{job['id']} until status == "done"

const res = await fetch("https://api.bettervideo.io/v1/jobs", {
  method: "POST",
  headers: { "Authorization": "Bearer YOUR_KEY", "Content-Type": "application/json" },
  body: JSON.stringify({ video_url: "https://.../clip.mp4", resolution: "1080p" })
});
const job = await res.json();

Privacy is the default, not an upgrade

• Never trained on. Pre-trained, fixed-weight models — your frames are enhanced and returned, never used to train AI.
• Never sold or shared. No third parties, no data brokers, no "anonymized" resale.
• Auto-deleted. Files self-delete (default 30 days), or delete any job instantly with one call.
• Proof on the Secure tier. Every deletion can return a signed deletion certificate, with a full audit log.