ChronoVerify vs Sightengine
Sightengine and ChronoVerify answer different questions about an image. Sightengine is a content-moderation API that returns probabilistic scores, including an AI-generated probability and a deepfake score. ChronoVerify reads and cryptographically validates content provenance and capture time and returns a deterministic verdict. Many teams run a detector and a provenance validator together. This page compares them factually, with every Sightengine claim linked to Sightengine's own pages and dated.
In one line each
- Sightengine returns a probabilistic AI-generated or deepfake score; ChronoVerify returns a deterministic provenance verdict. They answer different questions.
- ChronoVerify validates C2PA Content Credentials against the official trust list and reads capture time; Sightengine's core detection is pixel-based and ignores provenance metadata.
- Sightengine bills per operation, and an AI-generated-image check and a deepfake check each count as 5 operations; ChronoVerify is a flat $0.003 per image no matter how many checks run.
- ChronoVerify produces a signed PDF audit report. Both are self-serve.
At a glance
| What buyers weigh | Sightengine | ChronoVerify |
|---|---|---|
| Primary purpose | Content moderation and image, video, and text analysis, with AI-generated and deepfake detection as one product category.1 | Validate content provenance and capture time and return a deterministic verdict. |
| What a result is | A probabilistic score from 0 to 1 per model, for example an ai_generated confidence and a deepfake confidence, returned as JSON.23 | One verdict (provenance_confirmed, consistent, inconclusive, metadata_anomaly, or manipulation_indicated) plus a 0 to 100 confidence and every signal. |
| C2PA Content Credentials | Its core detection is purely pixel-based and ignores EXIF and C2PA provenance; it offers a separate free browser-based C2PA checker utility.26 | Reads and cryptographically validates embedded C2PA credentials against the official C2PA and CAI trust lists as a core part of every verdict. |
| Capture time | Its AI detection is pixel-based and ignores EXIF metadata, so it does not report a capture time.2 | Reads the EXIF capture time, notes its source, and checks it for internal consistency. |
| Pricing model | Per operation. A free tier of 2,000 operations per month, Starter $29 for 10,000, Pro $99 for 40,000, then $0.002 per additional operation. An AI-generated-image check and a deepfake check each count as 5 operations.45 | Flat $0.003 per image regardless of how many checks run, with no per-operation multipliers; the first 100 each month are free. |
| Auditable record | Returns JSON scores or an in-dashboard analysis; no signed audit artifact is described on its pages.2 | An Ed25519-signed PDF audit report at $2.00 per report. |
| Getting started | Self-serve sign-up with a free plan, subject to daily and monthly limits; the Enterprise tier is sales-gated.1 | Self-serve: a free public verifier, an API key in minutes, and prepaid credits from $5. |
Sources and check dates below.
Where each fits best
Choose Sightengine, or another probabilistic detector, when you want a likelihood score that an image is AI-generated or a deepfake, or broad content moderation across nudity, violence, and more.
Choose ChronoVerify when you need to read and cryptographically validate provenance and capture time, work with C2PA Content Credentials, and keep a signed audit record. It is provenance-first. See the method page and the pricing page.
What a verdict tells you
ChronoVerify returns the photo's capture time, the capture device, the validated provenance state, and the file hashes, as a typed verdict with a confidence. It is not a deepfake or AI-generation detector, and a verdict is investigative triage, not proof: a clean result means a file's saved data is internally consistent, not that the scene it shows is real. Never use a verdict as the sole basis for an automated decision about a person.
ChronoVerify's own pixel layer is classical forensics, not AI scoring: block-level error-level analysis and noise-dispersion signals flag locally edited or spliced regions as triage for human review, with documented failure modes on recompressed images.
An auditable record
For teams working toward EU AI Act Article 50 transparency, ChronoVerify reads and validates the C2PA Content Credentials the Commission's Code of Practice points to, and produces a signed audit record of each check. See the compliance page for scope. ChronoVerify is a verification and audit layer, not a certification, and not legal advice.
Common questions
Is ChronoVerify an AI image detector like Sightengine?
No. Sightengine returns a probability that an image is AI-generated. ChronoVerify is provenance-first: it validates Content Credentials and capture-time metadata and returns a deterministic verdict.
Does Sightengine check C2PA Content Credentials?
Its core detection is pixel-based and deliberately ignores C2PA provenance, and it offers a separate free browser-based C2PA checker utility. ChronoVerify validates C2PA credentials against the official C2PA and CAI trust lists as a core part of every verdict.
How does the pricing compare?
Sightengine bills per operation: a free tier of 2,000 operations per month, Starter at $29 for 10,000, and Pro at $99 for 40,000, plus $0.002 per additional operation, and an AI-generated-image check and a deepfake check each count as 5 operations. ChronoVerify is a flat $0.003 per image regardless of how many checks run, with the first 100 each month free.
Can I use both?
Yes. A common pattern is a probabilistic detector for an AI-or-real signal and ChronoVerify for provenance validation, capture time, and a signed audit record. They answer different questions.
Is ChronoVerify C2PA conformant?
Yes. ChronoVerify is a C2PA Conformant Validator on the C2PA Conforming Products List, record 019f8a20-6452-7a43-b11b-59d0b0e4a84a, covering validation of JPEG, PNG, WebP and AVIF under C2PA specification 2.2. The list is public, so you can check the record yourself. It covers validation rather than generation: ChronoVerify reads and validates Content Credentials, it does not sign them.
Sources
Every Sightengine claim on this page is taken from Sightengine's own public pages, linked below and checked on the date shown. Prices and product details change, so these claims are re-checked on a quarterly basis.
- Sightengine home: content moderation and image, video, and text analysis APIs (Sightengine page, verified June 29, 2026)
- Sightengine docs: AI-generated image detection, ai_generated confidence 0 to 1, detection is purely pixel-based, metadata, EXIF, and C2PA provenance ignored (Sightengine page, verified June 29, 2026)
- Sightengine docs: deepfake detection, confidence score between 0 and 1 (Sightengine page, verified June 29, 2026)
- Sightengine pricing: Free 2,000 operations per month, Starter $29 for 10,000, Pro $99 for 40,000, $0.002 per additional operation, Enterprise custom (Sightengine page, verified June 29, 2026)
- Sightengine FAQ: how operations are counted, AI-generated and deepfake detection cost 5 operations each (Sightengine page, verified June 29, 2026)
- Sightengine tutorial: a free online C2PA checker reads and validates a C2PA manifest, while AI detection analyzes the visual content without relying on metadata (Sightengine page, verified June 29, 2026)
See what it returns on one of your own photos.
Try the free verifier