RED-Alert · Project

AI-Powered Real-Time Detection of Terrorist Content Across Social Media in Multiple Languages

digitalTestedTRL 6

Imagine having a security guard who can read every social media post in multiple languages at once and instantly spot when someone is planning something dangerous — that's what RED-Alert built. The system watches online conversations, understands what people actually mean (not just keywords), maps who's connected to whom, and fires off alerts the moment something looks like a real threat. It was tested by 6 police agencies across Europe and designed so that ordinary people's privacy stays protected while only flagging genuinely suspicious activity.

By the numbers

Law enforcement agencies validated the system

Consortium partners across multiple countries

Countries represented in the consortium

Euros in requested EC funding

Months of project development

Total project deliverables completed

SMEs in the consortium

The business problem

What needed solving

Law enforcement agencies across Europe face an impossible task: monitoring vast volumes of social media content in dozens of languages for signs of terrorist activity, while respecting citizen privacy and acting fast enough to prevent attacks. Current tools rely on keyword matching that misses context, cannot map network connections between suspicious accounts, and cannot process the sheer volume of multilingual online content in real-time.

The solution

What was built

The project delivered an integrated platform combining natural language processing, social network analysis, AI, and complex event processing to detect terrorist content in real-time across social media. Concrete demo deliverables include hierarchical models for content classification and a data networked privacy tool, with 18 total deliverables completed over the project lifecycle.

Audience

Who needs this

National police forces and counter-terrorism units needing real-time social media monitoringIntelligence agencies tracking cross-border online radicalization networksSocial media platforms required to detect and remove extremist content under EU regulationsPrivate cybersecurity firms offering threat intelligence and online risk monitoring servicesInternational security organizations like Europol coordinating multi-country operations

Business applications

Who can put this to work

Public Safety & Law Enforcement

enterprise

Target: National police forces, intelligence agencies, and Europol-affiliated security organizations

If you are a law enforcement agency struggling to monitor terrorist recruitment and propaganda across social media platforms in multiple languages — this project developed an integrated system combining natural language processing, social network analysis, and complex event processing that was validated by 6 law enforcement agencies to detect threats in real-time while preserving citizen privacy.

Social Media & Online Platforms

enterprise

Target: Large social media companies and content moderation service providers

If you are a social media platform dealing with the pressure to remove extremist content quickly and accurately — this project developed AI-driven content analysis tools that turn unstructured social media data into structured events, enabling rule-based detection of dangerous content patterns that go beyond simple keyword matching.

Cybersecurity & Risk Intelligence

mid-size

Target: Private threat intelligence firms and corporate security consultancies

If you are a cybersecurity firm offering threat monitoring to enterprise clients and need to track radicalization signals across online networks — this project developed social network analysis and complex event processing tools that map connections between accounts and detect coordinated activity patterns in real-time.

Frequently asked

Quick answers

What would it cost to deploy this system?

The project was funded with a requested 5M Euros in EC funding across 36 months for development and validation by 18 partners. Licensing or deployment costs for end users are not specified in the available project data. Interested agencies should contact the coordinator for pricing.

Can this scale to monitor millions of social media posts?

The system was designed for real-time processing using Complex Event Processing (CEP) combined with NLP and social network analysis. It was validated by 6 law enforcement agencies, suggesting it handled operationally relevant data volumes. Exact throughput benchmarks are not published in the available data.

Who owns the IP and how can we license it?

The project was coordinated by Software Imagination and Vision SRL (Romania) with 18 consortium partners across 8 countries. IP ownership is typically shared among consortium members under Horizon 2020 rules. Licensing arrangements would need to be negotiated with the coordinator and relevant partners.

Which languages does the system support?

The objective states the system outperforms existing solutions in number of languages supported. With consortium partners from 8 countries (Spain, France, Hungary, Israel, Moldova, Malta, Romania, UK), the system likely covers major European languages. Based on available project data, the exact language list is not specified.

How does this protect citizen privacy?

Privacy preservation was a core design requirement. The project produced a dedicated 'Data networked privacy tool' as one of its demo deliverables. The system was validated by 6 law enforcement agencies with privacy compliance as a key performance criterion.

Is this ready for deployment today?

The project ended in September 2020 and delivered 18 deliverables including demo-level hierarchical models and a privacy tool. It was validated by 6 LEAs including Europol involvement. Based on available data, the technology reached a tested/validated stage but commercial deployment evidence is not documented.

Consortium

Who built it

The RED-Alert consortium brings together 18 partners from 8 countries with a strong operational focus: 7 industry partners (39% ratio) and 5 SMEs alongside 4 universities and 1 research center. The coordinator, Software Imagination and Vision SRL from Romania, is a private company. Notably, the consortium includes 6 law enforcement agencies as end-user validators and has Europol participation for dissemination, which gives the technology direct credibility with security buyers. The geographic spread across Spain, France, Hungary, Israel, Moldova, Malta, Romania, and the UK provides multilingual coverage essential for cross-border threat detection.

SOFTWARE IMAGINATION AND VISION SRLCoordinator · RO
SERVICIUL DE PROTECTIE SI PAZAparticipant · RO
BIRMINGHAM CITY UNIVERSITYparticipant · UK
MINISTRY OF PUBLIC SECURITYparticipant · IL
MAYOR'S OFFICE FOR POLICING AND CRIMEparticipant · UK
INFORMATION CATALYST SLthirdparty · ES
INFORMATION CATALYST FOR ENTERPRISE LTDparticipant · UK
SERVICIUL DE PROTECTIE SI PAZA DE STATparticipant · MD
CITY ST GEORGES UNIVERSITY OF LONDONparticipant · UK
EUROSOFT DEVELOPMENT SAparticipant · RO
EOTVOS LORAND TUDOMANYEGYETEMparticipant · HU
INSIKT INTELLIGENCE S.L.participant · ES
MINISTERIO DEL INTERIORparticipant · ES
MALTA INFORMATION TECHNOLOGY LAW ASSOCIATIONparticipant · MT
REICHMAN UNIVERSITYparticipant · IL
MINISTERE DE L'INTERIEURparticipant · FR

How to reach the team

Software Imagination and Vision SRL (Romania) — use SciTransfer's coordinator lookup service to find the right contact person

Order a report on this consortium See SOFTWARE IMAGINATION AND VISION SRL profile →

Next steps

Talk to the team behind this work.

Want to explore how RED-Alert's AI threat detection technology could fit your security operations? SciTransfer can connect you directly with the development team and arrange a technical briefing.

Order a report on this topic More in Digital & ICT