AI-generated image analysis report
Public reports turn detection into an indexable evidence archive. This example shows the structure ZeroTrue can use for AI-generated image analysis.
Report example
Image authenticity report
- Texture repetition
- Missing provenance
- Unusual edge transitions
- Face region anomalies
Search intent
Public report example for AI-generated image analysis
Primary evidence
Diffusion artifacts, Metadata gaps, Object boundary anomalies
Recommended action
Use confidence scores with source context, policy thresholds, and human review.
Report structure
A public report should summarize the media sample, score, indicators, metadata, and limitations in a stable URL.
- Media preview
- Confidence score
- Forensic indicators
- Metadata and provenance notes
Forensic indicators
Image reports should avoid vague claims and instead list observable evidence.
- Texture repetition
- Edge inconsistency
- Metadata gaps
- Face or object artifacts
Indexable evidence archive
Public reports can build topical authority when they contain unique examples and responsible explanations.
- Unique sample context
- Timestamped model output
- Linked methodology
- Human-readable explanation
Use cases
Newsroom verification
Marketplace fraud review
Brand impersonation checks
Sample report preview
Media preview
Safe sample, redacted upload, or generated demonstration asset.
Public reports should only expose media that is lawful, consented, and safe to publish.
Confidence
Synthetic likelihood: 91%
Texture repetition
Evidence item linked to score calibration, source context, and known uncertainty.
Missing provenance
Evidence item linked to score calibration, source context, and known uncertainty.
Unusual edge transitions
Evidence item linked to score calibration, source context, and known uncertainty.
Face region anomalies
Evidence item linked to score calibration, source context, and known uncertainty.
Evaluation table
| Criterion | What to check | Why it matters |
|---|---|---|
| Coverage | Text, image, audio, video, code. | Synthetic media risk rarely stays in one format. |
| Explainability | Score, indicators, timestamps, metadata, limitations. | Reviewers need evidence, not a black-box verdict. |
| Accuracy risk | False positives, false negatives, calibration. | High-impact workflows require documented uncertainty. |
| Workflow fit | API, batch, reports, retention, reviewer queues. | Search traffic must convert into a usable product path. |
Methodology and limitations
How to read the score
Detection output should be read as calibrated evidence. A high score means the observed signals are consistent with synthetic or manipulated media under the current model and sample conditions. It does not prove authorship, intent, or model attribution by itself.
Where review is required
Short samples, heavy editing, compression, translation, re-recording, mixed human-AI content, and unseen generators can reduce confidence. Use human review, source context, and policy thresholds before high-impact enforcement.
Next step
Match the action to the visitor intent: detector pages should lead to a scan, research pages to a downloadable report, enterprise pages to a demo, and developer pages to API keys or playground examples.
FAQ
Should every scan be public?
No. Public reports should only be created for samples that can be shared safely and legally.
Can reports be indexed?
Yes, public reports can be indexable when they contain unique evidence and do not expose private data.
What makes a report useful?
A useful report includes evidence, limitations, metadata, timestamps, and a clear confidence explanation.