Sys.Directive // 0x4A/Corporate Security

Enterprise Image Verification

Check images for AI generation and tampering before they reach your people. Send suspicious profile photos, documents and screenshots to a REST API and get a score, a verdict and evidence for your reviewers.

REST API
HOSTED IN GERMANY (EU)
IMAGES DELETED AFTER ANALYSIS
THREAT VECTOR

Synthetic Impersonation & Fake Images

Threat actors use generative AI to produce convincing faces, fabricated documents and edited screenshots to impersonate executives and suppliers. Sightova checks the image itself: submit a profile photo, an attachment or a still from a video call and see whether it was generated or edited.

  • Check a suspicious profile photo, ID photo or attachment and get an AI-probability score
  • Each scan looks for invisible watermarks (Google SynthID, OpenAI) and reads C2PA credentials and EXIF metadata. The model also reads the pixels, so detection works when metadata has been stripped
  • Call the REST API from your own workflow. You set the threshold and decide what happens next
EXAMPLE API REQUEST

POST /api/v1/detect/ai

X-API-Key: sk_<your-key>


{
  "request_id": "req_8f2a1c9e7b04",
  "endpoint": "/api/v1/detect/ai",
  "result": {
    "ai_probability": 0.97,
    "real_probability": 0.03,
    "confidence": 0.94,
    "verdict": "ai-generated"
  }
}

[OK] Score, verdict and evidence returned in seconds.

[OK] You set the threshold and decide what happens next.

_

Security & Integration Highlights

QUERY: SELECT * FROM secops_features
ENT-ID-01
SECURITY MODULE

AI-Generated Image Detection

Check profile photos, ID-card photos and other images for signs of AI generation. Each scan also looks for invisible watermarks (Google SynthID, OpenAI) and reads C2PA content credentials and EXIF metadata, so a watermark or credential can name the tool behind an image.

AI IMAGE DETECTIONSYNTHIDC2PAEXIF
ENT-DOC-02
SECURITY MODULE

Document Tampering Screening

Scan images of invoices, receipts and IDs for edited regions. You get a verdict and the share of the image that looks tampered, plus a heatmap that shows reviewers where to look on Starter and up.

INVOICE IMAGESTAMPERED REGIONSHEATMAPSTARTER+
ENT-KEY-03
SECURITY MODULE

Account & API Key Security

Passwords are hashed with bcrypt. API keys are shown once and stored only as hashes, and can be revoked from the dashboard at any time. Sign in with email and password or with Google.

BCRYPTHASHED API KEYSKEY REVOCATION
ENT-EU-04
SECURITY MODULE

EU Hosting in Germany

Sightova runs on Hetzner in Nuremberg, Germany, in ISO 27001 certified data centres. Web traffic passes through Cloudflare and is encrypted with TLS 1.2 or higher. Self-hosted deployment can be discussed under an enterprise agreement.

HETZNER NUREMBERGEU HOSTINGTLS 1.2+
ENT-DAT-05
SECURITY MODULE

Images Deleted After Analysis

An image is written to a temporary file during the request and deleted afterwards. Your images are not used for model training. Your scan results stay in your account history. The full data-handling detail is on the Trust Center.

DELETED AFTER REQUESTNO TRAININGRESULTS IN YOUR ACCOUNT
ENT-API-06
SECURITY MODULE

REST API Integration

Integrate from your own workflow with an API key, on every plan, and start free with no credit card. Four endpoints: AI image detection, NSFW, violence and document tampering. Send one image per request (JPEG, PNG, GIF or WebP, up to 50 MB) and get a full check back in seconds. An uptime commitment can be discussed under an enterprise agreement.

REST APIAPI KEY4 ENDPOINTSDOCS AT /DOCS
// SECURITY FACTS

Built for Security Review

The questions your security team asks first: where images are processed, how long they are kept, how keys are stored and how detection is tested. The full detail is on our Trust Center.

Read the Trust Center

Published Testing

Detection is tested on 250k+ benchmark images and 50+ generative models, and the methodology and results are published at /testing-methodology.

EU Hosting

Application, database and detection models all run in Germany. A GDPR data processing addendum applies to every account, with signed copies on request.

Transport Security

Connections use TLS 1.2 or higher. Card data is entered directly into Stripe-hosted fields and never reaches Sightova servers.

Security Questionnaires

We complete security questionnaires on request, so your review can start from straight answers.

Review Sightova With Your Security Team

Talk to us about volume pricing, security questionnaires and a signed DPA. Self-hosted deployment and an uptime commitment can be discussed under an enterprise agreement.