Imagine someone sends you a shocking video or news article — how do you know it's real? WeVerify built a toolkit that acts like a digital forensics lab for online content. It checks images, videos, and social media posts using AI to spot fakes, tracks how disinformation spreads across networks, and keeps a shared blockchain ledger of confirmed fakes so nobody falls for the same trick twice. Think of it as a "reverse image search on steroids" combined with a community of journalists who cross-check each other's findings.
Online disinformation and AI-generated fake content (deepfakes) are causing real financial and reputational damage to businesses, media organizations, and governments. Even experienced journalists struggle to verify whether images, videos, and social media posts are authentic. Companies need automated, scalable tools to detect fakes before they spread — but building this in-house requires AI expertise most organizations don't have.
The project delivered a full verification platform (reaching v3.0) with cross-modal content verification tools, a blockchain-based public database of confirmed fakes, disinformation flow analysis with an early warning system, and both open source and premium commercial versions designed for newsroom integration.
If you are a news organization dealing with the growing flood of manipulated images, deepfake videos, and fabricated social media posts — this project developed a cross-modal verification platform tested by journalists at Deutsche Welle and AFP. It integrates into existing content management systems and already had a community of more than 2,700 users through the InVID verification plugin predecessor.
If you are a brand protection team dealing with disinformation campaigns that damage your company's reputation — this project built social network analysis tools and a disinformation flow analysis system that tracks how false narratives spread online. The early warning system can detect fabricated content before it goes viral, giving your team time to respond.
If you are a government agency tasked with protecting democratic processes from online manipulation — this project delivered a blockchain-based public database of known fakes verified by at least 3 independent nodes. The platform supports human rights organizations and emergency response teams with micro-targeted debunking tools validated across 6 countries.
The WeVerify platform core is open source, designed to engage citizen journalists and communities at no cost. A premium version is also offered for newsrooms needing advanced features and integration with in-house content management systems. Specific pricing for the premium tier is not disclosed in the project data.
The platform went through multiple development cycles (v2.0 and v3.0) with dedicated scalability, throughput, and robustness evaluations documented in each release. It was validated by professional journalists at major organizations including Deutsche Welle and AFP, as well as a community of more than 2,700 users of the predecessor InVID plugin.
The core platform and algorithms are open source, meaning they can be freely used and adapted. The premium version with advanced newsroom integrations was developed by the 9-partner consortium led by Ontotext AD (Bulgaria). Licensing terms for premium features would need to be discussed with the consortium.
The system uses a distributed ledger where at least 3 partner nodes independently verify and record confirmed fakes. Hashing functions for text, images, and media are combined with the identity of the verification agent — such as a journalist or fact-checker — creating a tamper-proof record that any platform participant can check against.
The platform handles cross-modal verification — meaning it can analyze text, images, video, and social media posts together. It specifically addresses AI-generated synthetic multimedia content (deepfakes) using deep learning, and tracks disinformation flow across social networks with an early warning system.
The project closed in November 2021. Based on available project data, the open source components and the community of 2,700+ users from the InVID plugin suggest ongoing usage. The project website at weverify.eu and consortium partners like AFP and Deutsche Welle indicate continued interest, but current maintenance status should be confirmed directly.
The platform was developed under EU Horizon 2020 funding with a specific focus on democracy protection and human rights. Its participatory verification approach with human-in-the-loop machine learning aligns with EU principles on AI transparency. Specific compliance with the Digital Services Act or AI Act should be verified with the consortium.
The 9-partner consortium across 6 countries (Belgium, Bulgaria, Germany, Greece, France, UK) is strongly industry-oriented with 4 industrial partners and a 44% industry ratio. The coordinator Ontotext AD from Bulgaria is a private company specializing in knowledge graph and semantic technology. Major media partners include Deutsche Welle (Germany's international broadcaster) and Agence France-Presse, providing real-world validation environments. EU DisinfoLab adds policy and research credibility. With 1 SME and a EUR 2,499,450 budget, the consortium balances technical AI development with immediate deployment into working newsrooms — a strong indicator of market relevance.
Ontotext AD (Bulgaria) — semantic technology and knowledge graph company, reachable through weverify.eu
Want to integrate disinformation detection into your newsroom or brand protection workflow? SciTransfer can connect you with the WeVerify consortium partners who built and tested this platform with AFP and Deutsche Welle.
