AI projects

The AviaTor project is funded by the European Union and is currently in its final phase. The project started in 2019 and will end in 2024. The project aimed to build a prioritisation tool / database for law enforcement processing NCMEC reports (also known as industry reports or Cyber tipline referrals). The AviaTor database provides law enforcement with the tooling to prioritise these reports. AviaTor stands for Augmented Visual Intelligence and Targeted Online Research – meaning that the AviaTor database has the functionality to use visual intelligence as well as OSINT and hash matching to de-duplicate and prioritise reports. AviaTor is currently used by 19 national law enforcement agencies.

ZiuZ Forensic, Web-IQ, Timelex, DFKI, INHOPE

Camera Vision is a machine learning tool that identifies images that contain both nudity and a child.

Facebook

CameraForensics is an online portal allowing users to search for other images on the internet taken with a particular camera to support victim identification investigations. This tool also supports searching geographically for similar and identical images and for various metadata elements. All with the intent of helping investigators link covert and overt (public) personas on the internet. The tool is available via an online user interface and also via a number of products such as Griffeye Analyze, or as a fully integrated solution for agency or country-level databases (via API).

CameraForensics

Clearview AI acts as a search engine of publicly available images – now more than ten billion -- to support investigative and identification processes by providing for highly accurate facial recognition across all demographic groups. Similar to other search engines, which pull and compile publicly available data from across the Internet into an easily searchable universe, Clearview AI compiles only publicly available images from across the Internet into a proprietary image database to be used in combination with Clearview AI's facial recognition technology. When a Clearview AI user uploads an image, Clearview AI’s proprietary technology processes the image and returns links to publicly available images that contain faces similar to the person pictured in the uploaded image.

Clearview AI

The Content Safety API sorts through many images and prioritizes the most likely child sexual abuse material (CSAM) content for review. The classifier can target content that has not been previously confined as CSAM.

Google

Krunam provides breakthrough technology that identifies and classifies previously unknown CSAM images and video at scale. Our CSAM classifier protects your online community, your employees, your brand’s reputation, and, last but not least, millions of children around the world.

Krunam

Luxand's technology detects facial features quickly and reliably. The SOK processes an image, detect-s human faces within it, and returns the coordinates of 70 facial feature points including eyes, eye contours, eyebrows, Iip contours, nose tip, and so on.

Luxan

Instead of logging in to separate data systems, agents, detectives and investigators can conduct a single search for a suspect, target, or location and return data from all relevant systems. Palantir Gotham featres intuitive, user-friendly interface backed by powerful data integration software providing access to art available information related to an investigation in one place. Data is secured at a granular level (down to individual attributes describing each piece of data), ensuring that users can only see information for which they are authorized. Users can also collaborate on investigations and share information. Palantir Gotham can integrate data in any format, including existing case management systems, evidence management systems, arrest records, warrant data, subpoenaed data, record managements system (RSS) or other crime-reporting data, Computer- Aided Dispatch (CAD) data, federal repositories, gang intelligence, suspicious activity reports, Automated License Plate Reader (ALPR) data, and unstructured dat such as'document repositories and emails. Access restrictions can be applied broadly based on the data source, or granularly, down to individually securing attributes and metadata (e.g., a building's address or the timestamp associated with a photograph) These permissions govern how users access data All interactions, including searching, viking, and editing data, are recorded in a tamper-evident audit log.

Palantir

PredPol uses a machine-learning algorithm to calculate its predictions on where crimes will occur. Historical event datasets are used to train the algorithm for each new city (ideally 2 to 5 years of data). It then updates the algorithm each day with new events as they are received from the department. This information comes from the agency’s records management system (RMS). PredPol uses ONLY 3 data points – crime type, crime location, and crime date/time – to create its predictions. No personally identifiable information is ever used. No demographic, ethnic or socio-economic information is ever used. This eliminates the possibility for privacy or civil rights violations seen with other intelligence-led policing models.

Predpol

Qumodo Discover is a search engine for evidence. This technology uses AI to intelligently find connections within digital data at a web scale where users can query the system with an image, and it will find other images containing the same places, people, faces and objects. It has been tested with more than 100 million images and validated in a victim ID environment, with child sexual exploitation and abuse (CSEA). content

Qumodo

Safer is a child sexual abuse material (CSAM) detection, review, removal and reporting pipeline. It allows small and medium sized companies to have the same CSAM fighting tools as the largest ones.

Thorn

TraffickCam allows anyone with a smartphone to fight sex trafficking when they travel by uploading photos of hotel rooms to a law enforcement database. Photos uploaded to the free TraffickCam app are added to an enormous database of hotel room images. Federal, state and local law enforcement securely submit photos of hotel rooms used in the advertisement of sex trafficking victims to TraffickCam. Features such as patterns in the carpeting, furniture, room accessories and window views are matched against the database of traveler images to provide law enforcement with a list of potential hotels where the photo may have been taken to identify location of victims.

Exchange Initiative

Vigil At automatically detects, evaluates and categorizes child sexual abuse imagery. Thu system is capable of determining the severity of the sexual act in the image (Using the legacy UK 1-5 Category SAP Scale or the current UK 3 Categories A-C). The tool is available as part of Qumodo Classify, via a Cloud API or via a stand alone API for tools such as Griffeye Analyze. The tool scales linearly and can categorize millions of images per hour

Qumodo