Synthetic drugs like “bath salts” and weed performances no sign of slowing down. Other Governments try to identify and ban these substances, new ones are invented and brought to market faster — even if Overdose continue claim more lives.
Law enforcement agencies want to stay ahead of the curve and predict these drugs before they take off. So researchers at the University of British Columbia have created an AI model that can predict which new synthetic drugs are most likely to be produced and released on the market, helping police to collect Narrow drug investigations from months to days. The model is detailed in a New research published in Natural machine intelligence.
The synthetic drug industry is essentially an advanced form of chemistry. Underground chemists develop and play around with new molecules that mimic the psychoactive effects of more common drugs like marijuana. But the movement around these molecules means that drugs are not technically the same as illegal substances, which means users can get a high legal price and manufacturers don’t. may be prosecuted. The ability to create new synthetic drugs is fundamentally limited by the imagination.
But AI technologies have the uncanny ability to absorb large amounts of data and produce more realistic analytics. In this case, the researchers fed the AI model a database of hundreds of known psychoactive substances. The model learned to predict 8.9 million potential synthetic drugs could be developed — a testament to its ability.
The researchers then tested the AI against 196 synthetic drugs that hit the market after the model’s training began – so these were drugs they didn’t know existed. in. As it turns out, the AI predicted the arrival of more than 90% of those drugs.
Furthermore, the AI also learns to predict which types of molecules are more likely to appear in the illicit drug market than others. From a dataset of 196 new synthetic drugs, the model was able to accurately predict which chemical structure would make it into the top 10 most popular drugs with 72 percent accuracy. Improvements to the model with other bits of chemical data increased this accuracy to 86 percent.
According to the study authors, a number of authorities around the world have expressed interest in adopting and using this model as part of their investigations.
“There’s a whole world of ‘dark matter’ chemistry right now that’s out of our reach,” said study co-author and UBC medical student Michael Skinnider. in a statement. “I think there is a huge opportunity for the right AI tools to illuminate this unknown world of chemistry.”
https://www.thedailybeast.com/new-ai-tool-could-help-cops-predict-synthetic-drugs-before-they-hit-market?source=articles&via=rss | New AI tool can help police predict synthetic drugs before they hit the market