AI Tool Helps Law Enforcement Stop Human Trafficking
Emily Kennedy was 16 when she traveled to Macedonia. She was bewildered at the desperation in which children were cleaning the car she was riding inat a red light. Her friend told her that the kids were trafficked by the Russian mob, and would be punished if they didn’t bring enough money back.
This got Emily thinking. She continued onto college and was planning to get a law degree, but human trafficking stayed on her mind. Working on her senior thesis, she saw an opportunity with a few machine learning experts and engineers at the university who had created technology to solve other complex issues, such as epidemic detection and food safety.
She expressed her idea that this technology and online data could be used to fight human trafficking. Her passion pitching the idea coupled with the experts’ excitement was just the beginning.
The software, known as Traffic Jam, works by trawling online sex ads, of which a subset includes victims of human trafficking, featuring a surprising number of photos. The software uses Amazon Web Services’ facial recognition tool to match a victim’s photo with any online sex ads. Since traffickers move victims from city to city to avoid detection and capture, it’s imperative that the software moves quickly through thousands of ads.
Kennedy’s company, Marinus Analytics, has gathered online ads since 2011, helping detectives analyze characteristic ad misspellings to narrow down which may include trafficking victims. Another function is an image similarity search, “where a detective can basically upload a photo of a victim procured in a hotel room and see other photos of victims pictured in the same hotel,” Kennedy states.
The data also helps analyze event statistics where human trafficking takes place. Events like the Super Bowl show significant human trafficking activity, but other events have even more statistical significance, such as CES (the International Consumer Electronics Show), the Oracle Conference, and the Daytona 500.
The company has partnered with law enforcement officials, prosecutors’ offices, and a few nonprofits to help increase the impact on human trafficking. With an 88% success rate, Traffic Jam data has already led to hundreds of rescued human trafficking victims in North America.