AIR Fusion Verify
AIR Fusion Verify analyzes video, images, and audio for signs of AI generation and manipulation, right inside your existing ingest workflow.

@onscene
flooding downtown right now
*for illustration purposes only
When a story breaks, footage arrives from viewers and social platforms within minutes. Checking each file by hand slows your team down at exactly the wrong moment. Verify runs the analysis as media comes in, so your team spends its time on the files that need a closer look.
Why AIR Fusion Verify
What you get
Detection-only tools
A verdict and a confidence score, through a web tool or an API
AIR Fusion Verify
Five checks with the evidence behind every score, through the API or the workspace
Context
Detection-only tools
Looks for signs the file was AI-generated
AIR Fusion Verify
Checks for AI generation, plus frame, narrative and metadata analysis
Provenance and watermarks
Detection-only tools
Something to check separately
AIR Fusion Verify
Content credentials and watermark checks in the same request
Review
Detection-only tools
One reporter, one scan
AIR Fusion Verify
Your reviewers can add notes and record the final call*Workspace option only
The record
Detection-only tools
Copy the score into your own notes
AIR Fusion Verify
A full audit trail for every check and decision, ready for compliance
Infrastructure
Detection-only tools
Varies by vendor
AIR Fusion Verify
Runs at scale on Microsoft Azure, with regional hosting options
One request. Multiple checks. One set of findings.
Flags images and video that show signs of AI generation.
AI likelihood
0.95
Flags speech that sounds synthesized or altered.
Flags inconsistencies between footage and what's plausible.
Inspects provenance and editing history, when present in the file.
Signed by Runway AI, Inc. Nov 19, 2024
Checks for supported AI-generated content watermarks.
How it works
01
Send the media
Submit incoming media through the API or upload to AIR Fusion as they come in.
02
Act on the findings
Use scores and detection signals to clear low-risk files and route anything suspicious to a reviewer.
03
Record the decision
Journalists review the evidence and log their decision, through your own tools or the Verify review workspace. Every outcome is added to the audit trail.
AIR Fusion Verify is for newsrooms and journalists who need a quick way to verify incoming user-generated content for potential AI generation or manipulation.
No detection tool catches everything. AIR Fusion Verify gives your team the evidence to make an informed call, and never claims to be 100% certain.
AIR Fusion does not train AI models on your content. What happens in your workspace stays in your workspace, and the model providers we use are not allowed to train on it either.
Yes. AIR Fusion is built on Microsoft Azure and connects to your Microsoft 365 environment. Teams can get alerts in Teams, generate reports as Word documents, and store them in SharePoint. Enterprise plans include single sign-on.
Yes. Your files and metadata are stored on Microsoft Azure, encrypted in transit and at rest, and kept separate for your organization. Your contract specifies where that data is processed. If your newsroom needs content to stay in a particular country or region, talk to us about regional hosting.
AIR Fusion Verify supports common image, video, and audio formats, including MP4, JPEG, PNG, WAV, and MP3. You can upload up to 500 assets at a time. Contact us if you have specific file-size or format requirements.
Yes. Use the API to run checks inside your existing tools and workflows.
Give your newsroom more evidence before it publishes.
Talk to our team about evaluating AIR Fusion Verify with your own media and workflows.