Additional Technological Applications


Technologies used for anti-trafficking efforts need not be invented from scratch. In fact, a number of existing technologies can be repurposed for use by the anti-trafficking community. Examples of possible technological solutions include the following:

Information-Sharing Platforms for Anti-Trafficking Organizations

The need for information sharing is a common refrain among actors engaged in anti-trafficking efforts. In August 2010 and March 2011, the Annenberg Center on Communication Leadership & Policy sent research teams to the Mekong Subregion (Thailand, Cambodia, and Vietnam) to conduct a communication-needs assessment of organizations combating trafficking in persons. CCLP researchers met with leaders of more than 20 governmental, international, and nongovernmental organizations who indicated that the efforts of anti-trafficking actors often are hampered by the lack of effective inter-organization communication. These organizations have specialized expertise in trafficking issues and were proficient in their primary mission to assist victims. Yet the reasons behind the lack of information sharing are complex and include competing visions, values, missions, and funding sources, to name a few. Other challenges to an organization’s ability to share information include privacy concerns, victims’ rights, public safety, national laws, limited resources, and communication technology gaps.

A variety of Internet-based information-sharing platforms, from private social networking services to database-sharing software, could be developed to improve communications between actors and stakeholders. Other technologies can be developed to give victims and concerned members of the public more ways to communicate with service providers. One possible solution could involve developing a platform that captures calls from the trafficking reporting hotlines of various organizations and distributes those phone calls via text or email among those organizations. Data security and maintaining trust among organizations are among the challenges for innovators designing technological interventions to facilitate information sharing in this space.

The need to identify and improve information-sharing processes and technologies was highlighted by the Pacific Northwest National Laboratory in a recent report for the U.S. Department of Energy. The report found that information-sharing technologies were needed among federal agencies, regional anti-trafficking taskforces, and local police.1

Photo Recognition

While facial-recognition technology aims to match an individual’s countenance across different photos, photo-recognition technology addresses a slightly different issue—identifying copies of a particular photo among the sea of images on the Internet. In December 2009, Microsoft donated technology to the National Center for Missing & Exploited Children (NCMEC) to combat the spread of child pornography, in a move that illustrates recent advancements in the area of photo recognition, as well as the important role technology can play in supporting organizations such as NCMEC.

Microsoft’s PhotoDNA technology, developed in partnership with Dr. Hany Farid, a digital-imaging expert at Dartmouth College, makes it possible to locate copies of pornographic images of minors distributed online—even when copies had been digitally altered. Previously, the alteration of a picture made it difficult to match a photo to its original image, as the signature or “digital fingerprint” of the image was changed.

“The problem was that the signature was extremely fragile—the tiniest change to the image and the signature would be completely different,” said Farid. “The PhotoDNA technology extends the signature to make it robust and reliable, so that even if you change the image a little bit, we can still find it.”2

As a result, PhotoDNA creates a unique signature that remains consistent across copies of a photo—even when it has been digitally altered. Once known images are identified, NCMEC can share the signatures with online service providers, which can locate and remove copies of the photos.

“[PhotoDNA] is fast, it’s accurate, [and there are] no false alarms yet,” said Farid, noting that false alarms are a serious problem with facial-recognition technology.3

Microsoft’s PhotoDNA also addressed another challenge facing NCMEC—locating the worst forms of child pornography among the innumerable images online. “If I laid down in front of you a couple of billion images and asked you to hand me the ones that are inappropriate, you can imagine the scope of that problem,” said Farid. PhotoDNA, however, “can pluck out those inappropriate images from a sea of billions in a very fast, very reliable way.”4

Microsoft has adopted PhotoDNA for use in Hotmail, SkyDrive, and Bing. In May 2011, Facebook announced the adoption of PhotoDNA on its social networking site, applying the technology to all images uploaded to the site to help locate and remove images of child pornography. “Our hope and belief is that Facebook will be just the first of many companies to use the technology,” said Ernie Allen, president and CEO of the National Center for Missing & Exploited Children.5

Technology such as Microsoft’s PhotoDNA could conceivably be repurposed to assist with anti-trafficking efforts. “We think that this technology can also help disrupt the global sex trade,” observed Farid.6

Crowdsourcing and Flagging

Crowdsourcing technologies could enable the public to play an important role in anti-trafficking efforts. These technologies enable the public or a large defined or undefined group (the “crowd”) to send content that can be aggregated to produce possibly useful information.

A project called Survivors Connect created a program using the Ushahidi and Frontline SMS tools to map and connect international anti-trafficking organizations as well as survivors of trafficking.7According to Aashika Damodar of Survivors Connect, crowdsourcing must be deployed with an eye to flexibility, adapting procedures depending on the ultimate viewers of the aggregated data.8

For example, when collecting sensitive information about the location of shelters and support centers in unstable environments, Survivors Connect uses individualized passwords, allowing only verified humanitarian organizations to access the final maps. Through the organization’s Ayiti SMS SOS project in Haiti, text messages reporting instances of abuse or requests for services or advice are filtered through a referral and response team. The team, working in coordination with an array of NGOs and service providers, then responds to the messages or refers the information to other agencies. The data also is collected, stripped of sensitive information that may endanger the information provider, and plotted on a map showing instances of violence in a particular area.

While such examples of crowdsourcing have proven useful in the context of human rights issues, the method raises issues related to data collection that must be considered—namely, how to avoid overloading anti-trafficking organizations with information, how to organize information, and how to assess the veracity of reports. One possible method of making sense of the stream of information from a number of different sources may lie in tagging functionality. Tagging can allows users to classify the content of information found in classified advertisements, posts, images, or websites. A concentration of user-generated tags on a trafficking related category might increase the likelihood that reports relating to that area are accurate, prompting investigation into the circumstances.9

Social media sites frequently rely on users themselves to monitor the site through “flagging.” A review of the practices of 12 social networking sites 10revealed that all of the sites have some sort of reporting mechanism for users to flag inappropriate content. Craigslist requires more than one flag to affect a posting.11 More than 15% of all Craigslist postings are removed through community flagging, and approximately 98% of the posts removed are in violation of the site’s terms of use. 12

Flagging may have particular uses to address labor trafficking online. Systems could be designed for users to flag or rate advertisements on job-placement websites that have a history of labor or trafficking abuses. Again, verifying user-generated information or collected opinions is an issue to consider.

Mobile Phone Applications

Mobile and wireless technologies have been adopted more quickly than any communication technology in history.13Several anti-trafficking initiatives are harnessing mobile-phone penetration rates to educate consumers about human trafficking via mobile phone applications. In an example of technology used to create awareness of the economics of trafficking, Free2Work, a project that provides consumers with information about forced labor,14 distributes a mobile phone application that allows consumers to look up the ratings of companies and support companies demonstrating zero tolerance for forced labor.15
“The Free2Work application will provide conscious consumers with valuable information and company evaluations at the moment they need it most—when they shop,” according to Not For Sale Campaign president David Batstone.16

In June 2011, the U.S. Agency for International Development partnered with the Demi and Ashton Foundation and NetHope Inc. to announce the Stop Human Trafficking App Challenge, a contest to design the most innovative mobile-phone technology application to combat human trafficking in Russia.

“Traffickers in the region are increasingly using mobile technology to lure vulnerable people into modern-day slavery. The Stop Human Trafficking App Challenge makes mobile technology part of the solution,” according to Alison Padget, program manager for the NetHope anti-trafficking project.17

The applications will be judged according to various criteria, including: the potential for widespread application; usefulness in preventing human trafficking, increasing awareness, or providing services to victims; and functionality. Yet issues related to data security, building trusted systems, and the potential for these technologies to harm victims abound and should be a consideration in the development of mobile phone applications.

Notes

  1. Ibid. ^
  2. “New Technology Fights Child Porn by Tracking its ‘PhotoDNA,’” Microsoft News Center, December 15, 2009, http://www.microsoft.com/presspass/features/2009/dec09/12-15PhotoDNA.mspx. ^
  3. Hany Farid, professor of computer science at Dartmouth College, telephone interview with CCLP research staff, May 24, 2011. ^
  4. Ibid. ^
  5. Riva Richmond, “Facebook’s New Way to Combat Child Pornography,”  New York Times, May 19, 2011, http://gadgetwise.blogs.nytimes.com/2011/05/19/facebook-to-combat-child-porn-using-microsofts-technology/?partner=rss&emc=rss. ^
  6. Hany Farid, professor of computer science at Dartmouth College, telephone interview with CCLP research staff, May 24, 2011. ^
  7. “Freedom Datamap,” Survivors Connect, last accessed July 6, 2011, http://www.survivorsconnect.org/our-work/data-maps-crowdsourcing/datamap. A version 3.0 of the map is currently in the development process. For a more recent example of crowdsourcing and mapping, see “Slavery Map,” Not For Sale, last accessed July 13, 2011, http://www.slaverymap.org/. ^
  8. Aashika Damodar, Survivors Connect, telephone interview with CCLP research staff, July 20, 2011. ^
  9. As Meier notes, “An event that is reported by more than one source is more likely to have happened.” Patrick Philippe Meier, “Moving Forward with Swift River,” iRevolution, May 7, 2009, http://irevolution.net/2009/05/07/moving-forward-with-swift-river/. For information on the Ushahidi open-source crowdsourcing project, see http://www.ushahidi.com/, and for the Frontline SMS text-messaging tool, see http://www.frontlinesms.com/. For a discussion on the value of online collective intelligence, see Howard Rheingold, Smart Mobs: The Next Social Revolution (Cambridge: Perseus Publishing, 2002). For a counterpoint, see Jaron Lanier, You Are Not a Gadget: A Manifesto (New York: Alfred A. Knopf, 2010). ^
  10. These social networking sites include Facebook, Twitter, Myspace, Craigslist, Oodle, Backpage, Friendster, Google Groups, My Red Book, World Sex Guide, Eros, and My Provider Guide. ^
  11. “Thanks for flagging (or nominating for best of)!” Craigslist Online Community, last accessed June 27, 2011, http://sfbay.craigslist.org/flag/?flagCode=16&postingID=379321606. ^
  12. Ibid. ^
  13. Manuel Castells, Mireia Fernandez-Ardevol, Jack Linchuan Qiu, and Araba Sey, Mobile Communication and Society: A Global Perspective (Cambridge: MIT Press, 2007). ^
  14. “About Free2Work,” Free2Work, last accessed June 27, 2011, http://www.free2work.org/home. Created by the Not For Sale Campaign, Free2Work is jointly maintained by Not For Sale and the International Labor Rights Forum. The project is based on the idea of ethical consumerism, whereby consumers support the rights of workers producing the goods they purchase. ^
  15. “Development is underway to allow [Free2Work] app users to scan the bar code of any item, providing an instantaneous grade and data about that product.” The Free2Work mobile phone application is supported by Juniper Networks. “Smart Phone ‘App-tivism,’” Free2Work, last accessed July 29, 2011, http://www.free2work.org/app. ^
  16. “Smart Phone ‘App-tivism,’” Free2Work, last accessed July 29, 2011, http://www.free2work.org/app. ^
  17. NetHope, “NetHope, USAID and DNA Foundation Launch Mobile Trafficking App Contest in Russia and the Region,” press release, June 14, 2011, http://www.nethope.org/media/article/nethope-usaid-dna-foundation-launch-mobile-trafficking-app-contest/. ^