UPDATED 00:12 EDT / SEPTEMBER 07 2016

5611594783_8e9a533564_b NEWS

Algorithms to reduce child sex trafficking: an interview with the creators of Spotlight

Thousands upon thousands of escort ads are posted online every day in the United States, and unfortunately some of these ads represent and exploit children. It has been the job of law enforcement in the U.S. to look through these ads and detect the ones believed to be related to the exploitation of children and the trafficking of humans for sex. This is no small task considering the vast number of ads, but also because when the sifting is done manually, it is extremely time-consuming.

In 2014 Spotlight was launched by the nonprofit tech task force Thorn, along with co-founders Ashton Kutcher and Demi Moore. Spotlight is a web-based tool that leverages Amazon Web Services’ cloud computing service, machine learning algorithms and insight from officers in the field to connect disparate data, with a goal to improve the effectiveness and efficiency of domestic sex trafficking investigations.

The tool is currently in use by thousands of law enforcement agencies in the U.S. It has been an indispensable time-saver, according to law enforcement officials working with Thorn, and has already helped identify thousands of children advertised for sexual services on the Internet as well as both adult and child victims of human trafficking.

Ideal automation

SiliconANGLE recently got in touch with Thorn to talk about Spotlight. Brooke Istook, the company’s director of strategy, gave us the lowdown on the technology and the progress that has been made using it with US law enforcement.

Q: Could you give us a brief background concerning Thorn and why it was started?

Brooke Istook: Thorn was created in 2012 to understand and address the unique role technology played in child sexual exploitation. At the time there was no focused and coordinated effort across the tech industry, law enforcement and non-profit organizations to work together to address the issues and challenges technology presented. Thorn came in to address that gap and has continued to grow and make an impact in this space ever since.

Q: What exactly is Spotlight and how is it used?

Istook: Spotlight is Thorn’s human trafficking investigations tool. Its goal is to provide law enforcement with intelligence and leads about suspected human trafficking networks and individuals in order to identify more victims and connect them with resources. Spotlight was developed because we learned from Thorn’s nationwide survivor survey that 3 out of 4 child sex trafficking victims were advertised online in the course of their exploitation. We also learned from our research that over 150,000 escort ads are posted online each day across the US, and that human trafficking investigators had to manually sift through those ads to try to find child sex trafficking victims. Our goal with Spotlight was to use machine learning algorithms to more quickly sift through the thousands of ads posted each day and more quickly identify the highest risk profiles. We partnered with Digital Reasoning on development of this tool and launched Spotlight late 2014.

Q: Can you explain how such a tool works?

Istook: Spotlight a web-based application that connects data from various publicly available data sources that comprise the online commercial sex marketplace. The tool helps prioritize leads by leveraging machine learning algorithms that were built based on insight from officers in the field. The tool also culls through massive amounts of data in order to provide deep investigative tools to help officers understand the historical and full geographic reach of a victim’s trafficking situation. The objective of the tool is to improve the effectiveness and efficiency of domestic minor sex trafficking investigations and increase the number of children who are identified and connected with help resources.

Killing time

Q: Can you give an example of how it has been successful?

Istook: We know that Spotlight has been used in over 4400 investigations and been used to identify over 2300 victims of human trafficking, including over 900 children.  Spotlight has also been used to identify more than 1000 traffickers.

Q: Does the tool face any setbacks related to legislation, privacy issues?

Istook: No. Spotlight only uses publicly available data.

Q: Is there a possibility of inaccuracy? If not, could you explain its efficacy?

Istook: Spotlight’s power is in its ability to collate data to present investigative insight and in its algorithms which are meant to flag leads based on various risk indicators. The officer then considers the various risk indicators along with other data in the tool to decide the ultimately viability of the lead. And the officer then verifies each point of data throughout their investigation in order to ultimately identify a victim or trafficker. As with any environment that deals with messy data and predictive algorithms, there are chances of false positives, but this continues to improve over time.

Q: Going forward, how do you think machine learning tools such as this might be used? How might they evolve?

Istook: The intersection of child sexual abuse and technology often involves challenges related to data volume and scale. Machine learning and natural language processing will continue to be valuable tools in this space to deal with the massive amounts of data and help leverage it to identify and rescue more children more quickly.

Minority Report

Q: We hear a lot about “machine bias,” predictive technology getting the wrong man or woman. Can you assuage those fears?

Istook: Our algorithms are predictive in that they assign various levels of risk based on natural language processing. In our world, the human reviewer or investigator ultimately makes the final decision on a particular lead based on the combination of risk indicators plus other information outside the scope of the algorithms. Also, in order to stay relevant and accurate, the algorithms must continually be tweaked based on human input. We rely on the machine to help sift data and prioritize information and combine that with human input, judgment and intervention for ultimate decision making.

Q: Is this tool something law enforcement officials are aware exists? Is it now widely available in the US?

Istook: Spotlight is being offered for free to law enforcement and is currently being used in all 50 states by over 700 agencies and close to 3000 officers. We continue to work to get the word out to other officers so they are aware of the tool. We are also expanding into other countries in the coming year.

Photo credit: Ira Gelb via Flickrpred

Since you’re here …

Show your support for our mission with our one-click subscription to our YouTube channel (below). The more subscribers we have, the more YouTube will suggest relevant enterprise and emerging technology content to you. Thanks!

Support our mission:    >>>>>>  SUBSCRIBE NOW >>>>>>  to our YouTube channel.

… We’d also like to tell you about our mission and how you can help us fulfill it. SiliconANGLE Media Inc.’s business model is based on the intrinsic value of the content, not advertising. Unlike many online publications, we don’t have a paywall or run banner advertising, because we want to keep our journalism open, without influence or the need to chase traffic.The journalism, reporting and commentary on SiliconANGLE — along with live, unscripted video from our Silicon Valley studio and globe-trotting video teams at theCUBE — take a lot of hard work, time and money. Keeping the quality high requires the support of sponsors who are aligned with our vision of ad-free journalism content.

If you like the reporting, video interviews and other ad-free content here, please take a moment to check out a sample of the video content supported by our sponsors, tweet your support, and keep coming back to SiliconANGLE.