The tool, dubbed Söze, can reportedly help solve policing problems, with one of its benefits being the platform’s ability to discover avenues of potential interest faster than humans.

Specifically, Söze enables investigators and analysts to identify patterns across disparate data sets at a level of analysis not currently possible. According to its developers, Söze performs that analysis automatically.

Its developers suggest that during trials of Söze, the AI tool has been successfully proven on crime types such as terrorism, assassination, volume crime, organized crime, missing persons, human trafficking, homicide, financial fraud and scams, drug trade, cyber-crime and child exploitation, among others.

Söze is built on the Azure platform, and it promises to help alleviate repetitive analytical tasks that often eat analysts’ or investigators’ time and create a backlog within investigative departments.

Because the tool automates the so-called ‘bread and butter’ analysis common among investigative departments, the tool can free up analysts and investigators to focus on more specific areas of analysis and thus pursue individual investigative channels, the company explained.

Further, Akkodis suggests that Söze also uses machine learning and AI features to perform a deeper dive into existing data and unearth associations that would typically not have been possible. To accomplish this, Söze uses a combination of link and frequency analysis, communications analysis, network analysis, object discovery, vectoring, geo spatial analysis, text analysis and facial recognition.

During trials of Söze, its developers determined that the tool was able to review the evidential material in 27 cases in only 30 hours. This, developers estimated, would have taken up to 81 years for a human to do, thereby making the tool 23,600 times faster than humans performing detective work.

“I could imagine this sort of thing being really useful for cold case reviews. You might have a cold case review that just looks impossible because of the amount of material there and feed it into a system like this which can just ingest it, then give you an assessment of it. I can see that being really, really helpful,” its developers added.