You are currently viewing How AI Is Helping In The Addiction Fight

How AI Is Helping In The Addiction Fight

As we enter this new decade, expect to hear the term Artificial Intelligence (or AI) much more frequently. Machine learning, as it’s also called, will soon be used to automate the way we live our lives. It certainly removes the margin for human error and, interestingly enough, may soon be implemented in curbing national addictions. The Vox tech site Recode recently published a story on this exciting development, highlighting how AI is already being used to find illegal opioid sellers online.


With the support of the National Institute on Drug Abuse, artificial intelligence programmers are now implementing a tool that can reportedly track how digital drug dealers market and sell opioids. This can pertain to everything from online transactions to activities across the dark web.


The technology utilizes publicly available social media and internet data to create algorithms tied to illegal drug transactions. UC San Diego researcher Timothy Mackey is involved in the development of the tool and spoke to Recode about the intricate work involved behind-the-scenes.


“For a platform like Instagram, for example, where we see a lot of drug dealers, it’s a number of hashtags associated with different opioid communities, and then it’s usually information about how to contact the drug dealer and buy from them,” Mackey explained when outlining the development process.  “We use a type of artificial intelligence called deep learning to track down illegal drug sellers on an Instagram. This type of AI focuses on recognizing patterns in data, and in this case, in Instagram posts. The idea is to get an AI-based system to recognize what drug-selling content looks like so that it can automatically find new sale-related posts within a much larger set of internet content.”


Eventually the machine learning begins to set in, honing in on topics related to suspected drug sales (as opposed to general conversations about opioids). As the technology advances, a manual scrape of 30,000 social media discussions on a topic like fentanyl can be whittled down to 10 exact posts focusing specifically on illegal activity. This certainly has the potential to be groundbreaking, as online drug dealers are undoubtedly helping to fuel this crisis. If they could be caught and removed at a rapid rate, it would greatly reduce the availability of illicit painkillers.


Per Mackey, there is still a lot of work to be done before this AI algorithm is deemed fully functional. Regardless, we are certainly excited to see this type of prospect on the 2020 horizon.