What does this tool actually do?

It runs a series of automated checks on your uploaded image — examining metadata, compression patterns, resolution, and other file-level signals. It then produces a 0–100 risk score and a plain-language explanation. It is designed as a quick first screening step, not a forensic conclusion.

Is the free check really free?

Yes. You can upload one image (JPG, PNG, or WEBP, under 3 MB) and receive an instant automated result at no cost. The only paid option is the Human Review if you want a more careful second opinion.

What is included in the Human Review?

A trained analyst manually examines your image alongside the automated screening results. You receive a written summary by email within 7 working days. The fee is £149 per image.

Can this tool detect all deepfakes?

No tool can guarantee 100% accuracy. This is an initial screening layer that highlights common warning signs. Sophisticated manipulations may not be caught by automated checks alone, which is why we offer the Human Review upgrade.

What file formats are supported?

We accept JPG/JPEG, PNG, and WEBP image files. Maximum file size is 3 MB. We do not support video, audio, or document files at this time.

Is my uploaded image stored or shared?

We do not permanently store your uploaded images. Files are processed in memory and discarded after analysis. If you request a Human Review, the image may be temporarily retained for manual examination and then deleted.

What should I do if my image scores high?

A high score means the automated check found multiple warning signals. Consider requesting a Human Review for a more careful assessment. If you believe you are a victim of deepfake abuse, preserve evidence and consider contacting law enforcement or a legal professional.

Can I use the results in a legal case?

Our automated result is not a forensic report and should not be treated as legal evidence. The Human Review provides a more substantive written assessment, but for legal proceedings you would typically need a certified digital forensics expert.

Journalist's Guide: Verifying User-Submitted Photos & Deepfake Detection

In an era where AI-generated images can be created in seconds, newsrooms face an unprecedented verification crisis. User-submitted photos—once the backbone of citizen journalism—now require rigorous scrutiny before publication. This guide provides professional verification protocols for journalists, photo editors, and newsroom leaders.

Whether you’re verifying breaking news images from social media or archival photos from contributors, these standards protect your publication’s credibility and legal standing.

The Verification Crisis in Modern Journalism

The proliferation of deepfakes has transformed newsroom workflows. Consider these statistics:

AI image generation tools create over 20 million images daily
Deepfake incidents in news contexts increased 900% in 2024-2025
Legal liability for publishing manipulated images now extends to news organizations

Traditional verification methods—reverse image searches and source checking—remain essential but are no longer sufficient alone.

Pre-Publication Verification Protocol

Step 1: Source Authentication

Before technical analysis, establish provenance:

Contact the submitter directly via phone or video call
Request RAW files or original camera exports (not screenshots)
Verify the submitter was physically present at the claimed location/time
Check submitter’s history and credibility

Step 2: Technical Screening

Deepfake detection technical diagram showing AI analysis process

Run automated deepfake detection on all submitted images. Our newsroom screening tool provides:

Instant risk scoring (0-100 scale)
Technical analysis of compression patterns
Metadata examination
Written documentation for editorial records

Note: Automated screening is your first filter, not your final decision. High scores require escalation.

Step 3: Forensic Analysis (High-Risk Content)

For images showing elevated risk scores or breaking news sensitivity:

Request Human Review (£149 per image, 7-day turnaround)
Receive written forensic assessment suitable for legal documentation
Include analyst credentials for court proceedings if needed

Red Flags: When to Escalate Verification

Immediate escalation triggers for newsroom editors:

Political content: Images of candidates, protests, or government officials during election cycles
Crisis events: Natural disasters, conflicts, or accidents where verification affects public safety
Legal implications: Photos used as evidence or affecting active litigation
Anonymous sources: Submissions from unverified or first-time contributors
Viral potential: Images likely to spread rapidly across social platforms

Building Newsroom Verification Standards

Create Verification Tiers

Tier 1 (Automated): All user-submitted photos
Tier 2 (Enhanced): Breaking news + political content
Tier 3 (Forensic): Legal proceedings + high-stakes exclusives

Documentation Requirements

Maintain records for legal protection:

Screenshots of automated detection results
Correspondence with image submitters
Human Review reports for disputed content
Editorial decision rationale

Legal Considerations for Publishers

Publishing AI-generated images as factual content creates liability risks:

Defamation: Fake images showing individuals in compromising situations
Election interference: Politically motivated deepfakes published as news
Copyright issues: AI training data disputes affecting image rights

Demonstrating “reasonable verification efforts” protects against litigation. Our Human Review service provides documented expert analysis suitable for legal defense.

Implementation Checklist for Newsrooms

Immediate actions to implement today:

☐ Add deepfake screening to your submission workflow
☐ Train editorial staff on basic AI artifact recognition
☐ Establish budget for forensic verification (budget £150-300 per high-stakes image)
☐ Update publication guidelines to disclose verification methods
☐ Create public correction protocols for mistakenly published AI images

Partner With Verification Experts

Ban Deepfake offers newsroom partnership programs including:

API access for bulk screening
Priority Human Review turnaround (48-hour option)
Staff training sessions on deepfake detection
Documentation standards for legal compliance