Batch video checks
Select short MP4, MOV or M4V clips in your company workspace: up to 20 MB and 60 seconds per video. Review each result and download individual reports with a CSV summary.
Batch video checksAI video detector: check if a video is AI-generated
For a personal check, upload a clip up to 60 seconds and 200 MB in MP4, MOV, WebM, MKV, AVI or another supported video format. Review generated visuals, face manipulation and synthetic voice signals separately, then download the PDF report. Company batches and API uploads have their own limits.
Specialist checks Data handling explained Uncertainty shown
Check a video AI video detection for business uploads
Review uploaded listing clips, recorded messages or disputed footage. Use company batches for selected short videos or submit one API job per clip from your backend. This is file analysis, not live-call protection or identity verification.
AI image & video detection API
Connect the API to check video uploads in your application. Submit one job per video and track completion through job status or signed webhooks. API uploads have their own file limits.
AI image & video detection APIReports your team can review and reuse
Read the finding together with coverage, unavailable checks and limitations. Keep the same recorded result in JSON, text or PDF; company branding does not change the finding and an export does not rerun a paid check.
See the report fields and export optionsWhat does an AI video check cost?
Personal checks
The credit cost depends on the clip’s duration: 15 seconds uses 4 credits; 60 seconds uses 13 credits. The checker shows the cost before you start. Optional source research is charged separately.
View personal credit packsAPI and company batches
Company video checks cost €0.04 per started second, with a €0.50 minimum per completed video when billed in euros.
View API and batch pricingCompany billing covers completed checks, including inconclusive reports. Failed processing is not billed. Check the pricing page for supported currencies and the available options.
How to check an AI video without losing the context
Preserve the full clip
Keep the earliest copy with its audio and surrounding footage. Cropping, screen recording and re-encoding can remove details needed to interpret the result.
Read the available signals separately
Generated visuals, face manipulation and synthetic speech answer different questions. A clip without a usable face or voice cannot provide all the same evidence.
Check a suspected photo or video deepfakeVerify the event and speaker
Compare the original publication and trusted recordings. An uploaded clip check does not verify a live call, identify the speaker or confirm their request.
Inspect available Content CredentialsFrom an API key to a completed business check
Create a company workspace and key
Register as a business or developer and create your company API key. Browser batches use your signed-in company workspace and do not need an API integration.
Test the integration with fixed examples
Use sandbox fixtures to verify requests, job status and report handling. A fixture is a preset response, not an evaluation of your file or a detection-quality test.
Connect billing before real files
Add company billing, check the selected tariff and set a spending limit. Personal credits are separate. A completed inconclusive check is billable; failed processing is not.
Retrieve the report, then review the evidence
Follow each submitted job to completion by status or a signed webhook. Preserve its report and source context, and route missing or conflicting evidence to a reviewer instead of treating the score as an automatic decision.
Self-service activation; no sales call is required. Confirm API upload limits in the quickstart: personal checks, company browser batches and API uploads have different limits.
What this detector reviews
Generated visual patterns
Looks for frame-level patterns associated with synthetic video.
Deepfake face signals
Reviews visible faces for manipulation and face-swap indicators.
Synthetic speech
Checks the soundtrack when usable speech is present.
Reading the result
Use the score as a signal, not proof
The highest available model score informs the headline, while the individual checks remain visible. A low score means no strong signal was detected in this clip; it does not certify that every frame is authentic.
Important limitation: Very short clips, heavy compression, screen recordings, animation and videos without a clear face or voice can reduce reliability.
Read the full methodologySee how the evidence stays separate
Demonstration only: this is not a customer case, live result or performance claim.
Scenario
A reposted clip appears to show a real person making an urgent claim, but the original upload is unavailable.
Recorded finding
The report keeps generated-video, face-manipulation and synthetic-speech signals separate, including when they disagree.
What it means
A suspicious signal supports further review. It does not identify a creator or establish that the surrounding claim is false.
Next verification step
Locate the earliest full-length copy, compare key frames and speech with trusted sources, and verify the apparent speaker independently.
Complete sample
Inspect the report structure and its limitations
Inspect an illustrative AI video reportWhen this check can help
Viral clips
Check the earliest available recording and compare its caption with the original publication before reposting.
Recorded video messages
Review a suspicious message, then confirm the apparent sender through contact details you already trust.
Research
Keep the detector findings with the source history so another reviewer can see what remains unverified.
Questions people ask before checking
How can I check if a video is AI-generated?
Use the earliest, highest-quality clip available. Check its source and context, then use the AI video detector to review generated-visual, face-manipulation and synthetic-voice signals.
Can an AI video detector prove a clip is real?
No. A low score only means the tested models did not find a strong signal in that copy. Provenance, source history and the claim around the clip still need verification.
What is the difference between AI video and a deepfake?
AI video includes fully generated scenes, objects and motion. A deepfake more narrowly imitates or alters a real person, often through face replacement, lip-sync or synthetic voice. Use the dedicated deepfake video page when impersonation is the main concern.
Can I check a video without a face or voice?
A clip may still contain usable visual material. The report shows which checks were available; an unavailable face or speech result is not a finding that the clip is authentic.
Can an AI image check on one frame tell me whether the whole video is AI?
A frame can be checked as an image, but it covers only that still. It leaves out motion, other frames and speech. Use the video detector for a supported clip, then read which visual, face and voice checks were available.
Does this detector check live video calls?
No. It analyzes an uploaded recording, not an ongoing call. If you are allowed to process a recording, you can check that clip. Confirm sensitive requests through a separate trusted channel.
Continue beyond the detector score
Use source history, provenance and independent checks before making a high-stakes decision.
Read the AI-video verification guide