The bitcoin blockchain and artificial intelligence are being used to help parents of sex traffic victims locate their missing children.
Rebecca Portnoff at the University of California, Berkeley, and her colleagues have developed a tool to ferret out sex traffickers from legal advertisers using machine learning to spot common patterns in suspicious ads. The tool also uses public information from bitcoin, which is used to pay for the ads, to help determine who placed the ads, according to New Scientist.
Portnoff said sex traffickers use the internet to reach customers without having to reveal much information about themselves.
Carrie Pemberton Ford at the Cambridge Center for Applied Research and Human Trafficking said the tool will assist in investigations of sex traffickers and support prosecution in areas where funds move rapidly across financial instruments.
Sex Traffickers Use Backpage
About 4.5 million people have been forced into sexual exploitation. In the U.S., many are advertised on Backpage, the second largest classified advertising website. The typical sign that an advertisement involves trafficking is that the advertiser places numerous other ads on the site.
Normally, the tell-tale sign that an ad involves sex trafficking is that the person behind it is responsible for many other advertisements across the site.
AI Examines Writing Styles
Portnoff’s tool examines the way ads are written. Artificial intelligence highlights when similar styles are used.
Each bitcoin transaction gets logged on the blockchain, which does not store identities, but every bitcoin user has a wallet recorded next to their transaction. The AI tool searches the blockchain to identify the wallet corresponding to each advertisement.
Portnoff said it is easy to see when each ad was posted. She looks at the ad’s cost and time stamp, then ties it to a specific individual or group, giving the police a good candidate for investigation.
The tool narrows the number of ads that the police need to read for their investigations.
Tool Proves Effective
The researchers tested the tool on 10,000 advertisements in a four-week period, correctly identifying close to 90 percent of ads with the same author. The test yielded a “false positive” rate of only 1 percent. One bitcoin wallet was responsible for $150,000 worth of sex advertisements.
Backpage did not respond to New Scientist’s requests for comment at the time of this report. Portnoff’s team is working with various police departments and non-government organizations.
One trafficker who drugged and beat a 13-year-old girl was caught and sentenced to five years in prison.
Featured image from Shutterstock.