Something good can indeed come from crowdsourcing. A new project involving online volunteers aims to teach artificial intelligence to pinpoint locations where modern-day slavery is taking place.
Slavery, unfortunately, isn't yet a thing of the past. South Asia, in particular, is home to brick kilns that benefit from forced labor. Millions of people work in brick kilns in the region, and 70% of these people are basically slaves. Many of these people are working to pay off debts. There are 20,000 to 50,000 brick kilns in South Asia, forming the so-called “Brick Belt” across countries like Nepal, India, and Pakistan. These numbers, however, are just estimates. There are no official figures on how many brick kilns there are in the region.
New technology may be able to give authorities an exact number, or at least a closer estimate of how many brick kilns are present in the region. Volunteers are now working to teach an artificial intelligence program how to automatically recognize brick kilns from satellite images. As of now, volunteers have already identified 4000 possible slavery sites.
Volunteers analyze images from Google Maps and click on points that indicate brick kilns. The kilns have a distinctive look to them, which makes them easy to identify. The AI will then learn how to automatically identify brick kilns through a machine learning algorithm.
Kevin Bales, the project's leader, says that a lot of slavery sites are visible from space. You just have to know what to look for, and you'll find it. However, some sites are more difficult to spot than others. For example, open pit mines may be more difficult to spot than brick kilns. Open pit mines in countries like the Democratic Republic of the Congo are also likely to be using forced labor. Unlike brick kilns, however, open pit mines are harder to identify. These pits mostly just look like holes in the ground, unlike brick kilns that have a specific shape and colors.
Of course, identifying the sites where modern-day slavery is taking place is just one step. Not everyone who is working in a brick kiln is unwillingly working there. Authorities would need to physically visit the sites to see what's going on and how many people are being forced into working. With the help of the machine learning algorithm, however, authorities will know exactly where to look.
Image recognition software can also be vital in cases where satellite images won't be of much help. For example, image recognition can also help stop sex trafficking that takes place in hotels. Volunteers can use an app called TraffickCam to upload images of hotel rooms they visit. Sex traffickers often take photos of their victims in hotel rooms. The app can then compare photos from users to the photos that traffickers post. From there, the app can identify which hotels the victims were in.
With these advancements in image technology, society is getting closer to eradicating this pernicious stain on human history. Sooner rather than later, modern-day slavery may soon turn into a thing of the past.
Get weekly science updates in your inbox!