Recorded source
demo-portrait.jpg
640 × 640 pixels · 63,578 bytes
Inside a report
Sample AI image and video detection reports
See how image and video findings become a reviewable record. Inspect the field definitions, then download one image example in JSON, text and PDF. These are fictional examples, not customer files or measured test results.
Five questions to ask about a report
- Was the source suitable?
- Which checks actually ran?
- What signals were found?
- What could make them wrong?
- What should a person do next?
REPORT FIELD REFERENCE
What each part of the report tells you.
The same versioned structure is used for image and video reports. Available evidence depends on the submitted file and the selected analysis mode; API field names remain unchanged across languages.
| JSON field | What to read | Interpretation limit |
|---|---|---|
reportId · schemaVersion · generatedAt | Identify the recorded analysis, its format and analysis time. | A report ID is a reference, not a digital signature. PDF issue time may be later than analysis time. |
source | Names the submitted file and records available type, size and SHA-256. | A hash identifies exact file bytes. It does not prove the capture date, creator or truth of a scene. Missing values remain unavailable. |
analysis | Records the selected mode, completion status and credit use. | A completed report can still be inconclusive. Completion does not mean every possible check was available. |
verdict.code · verdict.aiLikelihood | Keeps the recorded category and detector signal together. | The number is not detector accuracy, fraud probability or confidence that the whole file is fake. A missing score is not zero. |
evidencePath[].signals | Separates model findings, file observations, provenance and any requested source research. | Read each signal label and availability. Confidence in an authentic class is not an AI-generation score; signals are not averaged or added. |
limitations · recommendedActions | Records gaps in the analysis and the next verification steps. | A low score or matching source cannot replace independent source, identity or claim verification. |
privacy | Describes the handling and visibility recorded for this report. | This fictional response-only example is not a retention setting for your account. Check the terms for the workflow you use. |
Use JSON, text or PDF in your own product · Open the complete JSON Schema
DOWNLOADABLE DEMONSTRATION
Inspect one image report in three formats.
JSON, text and PDF below use the same illustrative image record. The portrait is a published demo asset; its scores and findings are fictional examples, not measurements of the portrait or evidence of detection accuracy.
Recorded analysis
DFP-SAMPLE-IMAGE-0001
Mode: quick
This demonstration uses a fixed example analysis time.
File fingerprint
The JSON and PDF record the SHA-256 of the bundled portrait. It identifies the file used for the illustration; it does not establish authenticity.
Show recorded SHA-256
55c824c6118dfb18ef5156c3e7c6b5bf4cf9cd53ef44c3f1b5f5efa27c45910eInspect the complete illustrative image JSON
{
"schemaVersion": "deepfakepolicy.report.v1",
"reportId": "DFP-SAMPLE-IMAGE-0001",
"generatedAt": "2026-09-13T12:00:00.000Z",
"source": {
"modality": "image",
"sha256": "55c824c6118dfb18ef5156c3e7c6b5bf4cf9cd53ef44c3f1b5f5efa27c45910e",
"name": "demo-portrait.jpg",
"mediaType": "image/jpeg",
"format": "JPG",
"url": null,
"sizeBytes": 63578,
"pageCount": null,
"characterCount": null
},
"analysis": {
"mode": "quick",
"status": "complete",
"creditsUsed": 1,
"method": "DeepfakePolicy evidence synthesis"
},
"evidencePath": [
{
"id": "provenance",
"label": "Provenance",
"status": "limited",
"summary": "No verified origin credential or complete chain of custody was available for this report.",
"signals": []
},
{
"id": "metadata",
"label": "Metadata",
"status": "reviewed",
"summary": "Available format, coverage and metadata observations were reviewed.",
"signals": [
{
"label": "Submitted format",
"finding": "JPG",
"score": null,
"scoreLabel": null
}
]
},
{
"id": "media-signals",
"label": "Media signals",
"status": "not-available",
"summary": "No separate human-readable media observation was available for this modality.",
"signals": []
},
{
"id": "model-signals",
"label": "Model signals",
"status": "reviewed",
"summary": "The returned classifications are listed separately, including unavailable checks and any segment findings.",
"signals": [
{
"label": "AI-generated image",
"finding": "Indicators found",
"score": 82,
"scoreLabel": "82% signal score"
},
{
"label": "Deepfake face manipulation",
"finding": "No strong signal in this example",
"score": 6,
"scoreLabel": "6% signal score"
}
]
},
{
"id": "web-evidence",
"label": "Public web evidence",
"status": "not-available",
"summary": "Automated public-source research was not included in this analysis.",
"signals": []
}
],
"verdict": {
"code": "suspicious",
"label": "Suspicious signals detected",
"aiLikelihood": 82,
"confidenceBand": "not-rated",
"summary": "The analysed media contains signals that require review. The individual findings identify what to examine.",
"rationale": [
"The AI-generation detector returned an elevated signal.",
"No independently verified capture history is supplied in this example."
]
},
"recommendedActions": [
"Preserve the earliest available, highest-quality original and its surrounding context.",
"Confirm the publisher, identities and important claims through an independent channel.",
"Compare key frames, audio segments or images with an attributable original source.",
"Escalate high-stakes use to a qualified forensic reviewer before acting."
],
"limitations": [
"This is a probabilistic assessment, not proof of authenticity, authorship or intent.",
"Compression, editing, transcription, translation, short samples and unseen generation methods can change results.",
"A low score does not certify authenticity, and a high score does not identify a creator.",
"Public-source research can miss private, deleted, paywalled or unindexed material."
],
"privacy": {
"originalStoredWithReport": false,
"researchContributionSeparate": true,
"reportVisibility": "response-only",
"statement": "The report contains normalized findings, not the submitted source. Any optional research contribution is governed and deleted separately."
}
}Downloading these examples does not submit a file or run an analysis. Use the API quickstart to test the request flow with sandbox fixtures, or check a batch of your own media.
DFP-SAMPLE-IMAGE-0001DEEPFAKEPOLICY.REPORT.V1
AI IMAGEIndicators found — further verification required
AI image detector sample report
Evidence reviewed AI-generated demo portrait of a fictional person JPG image, 640 × 640 pixels Example findings; no detector measurement is claimed
Important limitationResizing, filters, screenshots and recompression can introduce or hide patterns used by an image detector.
NEXT ACTION
Locate the earliest available copy and ask the claimed publisher for the original file.
DFP-SAMPLE-VIDEO-0001DEEPFAKEPOLICY.REPORT.V1
AI VIDEO / DEEPFAKEMixed indicators — source confirmation needed
AI video and deepfake sample report
Evidence reviewed Short MP4 clip suitable for a first-pass review Reposted copy; original container and upload history unavailable Usable face and speech intervals present
Important limitationShort duration, screen recording and social-platform compression can reduce or distort detectable signals.
NEXT ACTION
Request the earliest full-length file and preserve the post URL, account name and timestamp.
Video findings use the same record structure.
Generated-video, face-manipulation and synthetic-speech findings stay separate. The video walkthrough explains their limits; its scenario has no downloadable source clip or measured scores. Inspect the video example.
Existing voice-report example
The illustrative voice report remains available for reference. Live business API processing supports images and videos; a legacy report example does not add a supported live modality.
Sometimes a file cannot be assessed reliably.
A short, damaged or heavily compressed file may not contain enough information. In that case, the report explains what is missing.
Need to ask about a report?
The report ID identifies the exported record and its format. Screenshots may be incomplete or altered; request the original export, then quote its report ID when asking DeepfakePolicy to review or correct a result.