Each year, human trafficking generates more than $150 billion in profits – at the expense of human life – with children accounting for around a third of its victims. It’s a practice that operates underground in almost every country on the planet, and despite the resources thrown at it – by law enforcement, non-governmental organizations, social media campaigns – it only seems to grow.
Experts on the topic say the tools used in even in the most developed societies fall far short of what is needed to put a dent in this grim and growing enterprise.
But artificial intelligence (AI) and machine learning (ML)? Those could change the game.
“Anytime you have to ingest large amounts of data and information, you try to identify trends and patterns, and it can be very difficult to do well,” Alma Angotti, managing director at Navigant Consulting, Inc. and former U.S. regulation official for the Securities and Exchange Commission, and the Financial Industry Regulatory Authority told Fox News. “Typically, it has been a rules-based system – like flagging transactions over a certain amount such or with a certain amount of frequency. The problem with that is you can’t identify patterns and problems.”
The U.S. Department of Homeland Security defines trafficking, also referred to as modern-day slavery, as a crime that “involves the use of force, fraud,or coercion to obtain some type of labor or commercial sex act.”
AI and ML, Angotti said, have the power to analyze more than just financial activity.
“It can highlight social, economic and even political conditions from hundreds of thousands of sources,” she said. “For example, law enforcement can look at young women of a certain age entering the country from certain high-risk jurisdictions. Marry that up with social media and young people missing from home, or people associated with a false employment agency or who think they are getting a nanny job, and you start to develop a complete picture. And the information can be brought up all at once, rather than an analyst having to go through the Dark Web.”
The current anti-money laundering (AML) environment, Angotti pointed out, relies on “simple transaction-monitoring rules to detect human trafficking," and it "simply does not have the capacity to consider, weight and examine the necessary number of inputs.”
“The problem with it now is that it produces a lot of false positives, so the real issues are lost in the weeds. If you use more machine learning, you can program more variables and machines can use the information it has and then teach itself how to better identify patterns,” she said.
“You get better alerts, and alerts that are more likely to recognize real risks.”
Despite the potential, for now those tools remain underused, Angotti said.
“The government could increase the leveraging of technology to support trained law enforcement in tackling this complex issue. It would be helpful to match data holders – government and financial institutions – with computational science and AI partners to provide deployable tools to identify human trafficking activity and the money flows associated with it,” she said. “The same networks that traffickers use to recruit their victims can also be used as the source of data to detect criminal activity.”