Active Computing & AI Mental Health

Identity Attributes Matrix Inititative (IAMI)

In plain English

AI plain-English summary

Police and intelligence agencies across Europe currently struggle to link fragmented identity data—fingerprints, facial images, online profiles, and alias records—across different databases and national borders. IAMI tackles this fragmentation directly. Existing identification tools work well for single, high-quality fingerprints or mugshots, but fail when faced with blurry CCTV footage, partial voice recordings, or fake social-media accounts. They also cannot rapidly cross-reference dozens of suspects simultaneously or flag bots and avatars as non-human. This slows investigations and leaves gaps that terrorists and organised criminals exploit. The project builds a three-dimensional "Identity-Attributes Matrix"—a software system that integrates biometric data, non-biometric clues (such as writing style or location patterns), and analytic results into a single searchable structure. It connects directly to existing case-management tools used by EUROPOL and INTERPOL. If successful, IAMI would allow an analyst to upload a corrupted passport photo and a suspicious username, and within seconds receive a ranked list of possible matches—including confidence levels for noisy data—while automatically filtering out fake identities. It also establishes a shared EU repository for terrorist threat assessment, enabling agencies to collaborate without duplicating effort. The result is faster, more accurate identification across borders, reducing false positives that waste investigative time.

View original technical description
AMI aims to revolutionize entity identification and resolution in security, intelligence, and investigation contexts. It addresses the challenges and limitations faced by current identification techniques and offers a visionary solution through advanced AI-powered software. The project's long-term vision is to propel EU LEAs and security/intelligence organizations into a new era of intelligence, investigation, and forensic capabilities. IAMI's core innovation is the development of an AI-powered software system that can seamlessly integrate with existing case management systems and analytic tools. At the heart of IAMI is the concept of the 'Identity-Attributes Matrix (IAM)', a 3D construct that encompasses a broad spectrum of multi-modal identity-related attributes, including biometric data, non-biometric identity-related data, and analytic results. These attributes are used to facilitate large-scale entity identification/resolution, including a broader range of entity types, simultaneous and rapid multi-entity identification/resolution, systematic and continuous analysis of flux of probe attributes, confidence in noisy/corrupted probe data, reduced false positive rates, access to contact attributes, and the ability to classify fake identities and bots/avatars. Furthermore, IAMI sets the groundwork for effective collaboration and data sharing across European agencies, including international organizationslike EUROPOL and INTERPOL through the establishment of a new EU Wide IAMI enrolled data collection repository for terrorist threat assessment and awareness which enriches their capabilities and assets. The realization of IAMI's vision will be achieved through the demonstration of the developed IAMI solution. It will be communicated and disseminated among EU LEAs and security/intelligence agencies through project activities, including deliverables, training curricula, workshops, and pilots. IAMI represents a significant step forward in EU fight against terrorism.

View the original record at the funder ↗

Related Research

Grants with similar aims, by meaning.

Identity Attributes Matrix Initiative
Attribution – data- analysis – countermeasures – interoperability: Adac.io
Immerse. Interact. Investigate
reliAble euRopean Identity EcoSystem
SID: An Exploration of Super-Identity

Original classification

EU-Funded

Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.