Technology
Technology OverviewImage Trust LayerDocument Trust LayerAudio Trust LayerInformation Trust LayerSource VerificationProducts
Fraud ScannerSafeCamSafeMicGardioIndustries
InsuranceBanking & KYCAuthoritiesMedia & PressHR & ExpensesDefenseResources
BlogCase StudiesAPI DocsPress AreaCompany
About UsJobsContactNewsrooms and fact-checkers receive images daily from sources that cannot be clearly verified. The VAARHAFT Fraud Scanner checks every image in seconds for manipulation, AI generation, and provenance. Giving you a documented basis for your decision.
.webp&w=1080&q=75&dpl=dpl_UCTwikveTDJ5bPjn2FpCxK3YLrqa)
AI-generatedKnown from
AI-generated images and manipulated content are fuelling disinformation at an unprecedented scale. Newsrooms that can systematically detect fakes and certify authentic material don't just protect their own credibility, they actively counter the spread of fake news.
Publicly available tools can create photorealistic images of events, places, and people that never existed. These images fuel fake news narratives and erode public trust in journalism.
Photos from conflict zones or protests are altered to tell a different story. Often, small edits are enough to completely reverse the context and create misleading narratives.
Images arrive via social media, anonymous tipsters, or agencies without full provenance data. Manual verification takes time that daily operations rarely allow.
Publish too late, you lose. Publish a fake image, you lose your readers' trust. Both are real problems.
Disinformation campaigns use synthetic media at industrial scale. Media organisations that cannot detect fakes risk becoming unwitting amplifiers of false narratives.
Readers increasingly question the authenticity of visual media. Newsrooms need a transparent, verifiable way to prove that their published content is genuine.
VAARHAFT offers two tools that work together: one for detecting fakes in incoming material, one for securing your own photographs with a cryptographic trust seal. Together, they enable newsrooms to only distribute verified, trustworthy content.
The Fraud Scanner analyses images and documents for AI generation, pixel manipulation, metadata anomalies, and duplicates. Results arrive in seconds with specific indicators of suspicious image regions.
For: Fact-checkers, photo editors, agency desks
SafeCam is a camera app that creates a cryptographic authenticity certificate at the moment of capture. This certificate proves the image was not altered after it was taken.
For: Photojournalists, foreign correspondents, newsrooms
The Fraud Scanner analyses incoming material. SafeCam secures your own photographs from the moment of capture. Together, they cover the entire editorial image pipeline.
Want to see how it works in your newsroom?
Book a DemoMedia outlets and press agencies are among the most exposed organizations when it comes to manipulated imagery. VAARHAFT sits inside your editorial workflow and flags problematic images before they reach publication.
Newsrooms receive hundreds of reader-submitted images during breaking news events. When the source is unknown or cannot be verified, fact-checkers run the image through the Fraud Scanner web tool in seconds, checking whether it has appeared elsewhere on the internet before or has been submitted multiple times from different accounts. The Fraud Scanner becomes a standard part of the fact-checking team's toolset, sitting alongside existing verification tools.
AI-generatedA complete analysis takes under five seconds. Fast enough for breaking news decisions.
No blanket verdict. Heatmap, metadata report, and duplicate search deliver concrete evidence that can be documented and shared.
AI manipulation methods evolve rapidly. VAARHAFT continuously updates its detection models without requiring any action from customers.
Images are deleted immediately after analysis. They are not stored and not used for model training.
Whether as a drag-and-drop web tool or via API into an existing system: integration requires minimal effort.
VAARHAFT was born from concrete industry needs. Features reflect real requirements, not theoretical assumptions.
VAARHAFT can be embedded under your own brand. Verification results appear as part of your editorial workflow.
From single-image verification to thousands of images per day. The infrastructure scales with your needs.
Deepfake scandals involving politicians and public figures are making headlines worldwide. Gardio scans the internet for unauthorised uses of a person's face and identity and raises the alarm before reputational damage is done.
Drag and drop images or documents directly into the VAARHAFT web tool. No setup, no IT required, ready to use immediately.
VAARHAFT checks the image for AI generation, manipulation, duplicates, and metadata anomalies. Full report with heatmap in under five seconds.
The analysis report shows exactly what was found: heatmap, metadata details, duplicate matches. Export as PDF for internal documentation.
The newsroom decides based on documented evidence. Certified images can be published with a VAARHAFT verification badge. For high-volume workflows, an optional API integration is also available.
TECHNICAL DETAILS
ITC DIA Amsterdam
2024
Versicherungsforen Leipzig
2024
Insurtech Hub Munich
2024
insureNXT · InsurLab Germany
2024
Versicherungsforen Leipzig
2023
JPEG, PNG, TIFF, HEIC, and PDF. Multiple images can be submitted as a ZIP package. Results are returned individually for each file.
Single images are analysed in under five seconds. Larger PDFs or batch submissions may take slightly longer but are designed for real-time workflows.
Yes. The system detects both fully AI-generated images and targeted pixel-level edits: altered backgrounds, removed objects, adjusted regions. The heatmap shows exactly which image region is suspicious.
No. Images are briefly processed on our servers for analysis and immediately deleted afterwards. Duplicate detection uses an anonymised hash algorithm that requires no permanent storage.
The Fraud Scanner delivers a statistical assessment, not a legally binding expert opinion. However, heatmap, metadata report, and duplicate search results provide a documented basis for editorial decisions and internal escalation processes.
SafeCam is a smartphone camera app. Photographers use it in the field like their regular camera app. Every shot automatically receives an authenticity certificate proving time, device, and immutability of the image.
Yes. The Fraud Scanner is available via REST API with structured JSON output and integrates with common CMS, DAM, or editorial systems. VAARHAFT provides integration documentation and onboarding support.
Yes. Images are not stored and not used for model training. Processing is fully data-protection compliant.
Talk to our team and learn how VAARHAFT fits into your editorial workflow. For individual editors as well as large agencies.
