Skip to content

Weaponizing AI: ML Model Creates 40,000 New Chemical Weapons in Six Hours

Several chemical mixes were more deadly than the extremely toxic VX nerve agent

Using AI for drug discovery is an area where a growing number of organizations are looking to do good. But a new study suggests a darker side: AI tools could be used to develop chemical weapons.

Collaborations Pharmaceuticals retooled an AI drug discovery system to successfully identify 40,000 new potential chemical weapons in just six hours, according to research that was published in the Nature journal and presented at a recent conference hosted by the Swiss Federal Institute for nuclear, biological and chemical protection.

The researchers wrote that the results “ultimately evolved into a computational proof of concept for making biochemical weapons.”

Collaborations’ scientists created a machine learning model that could identify molecular structures related to toxicity. They managed to generate several compounds more toxic than VX, a deadly toxic nerve agent used in the assassination of North Korean leader Kim Jong-un’s half-brother in 2017.

The pharma firm said the results of the rest were a “wake-up call” for those in the AI drug discovery community.

Its results suggest that toxicity datasets that provide a baseline model for predictions related to human health are readily available.

Despite their work focusing on VX-like compounds, the researchers warn the concept was “equally applicable to other toxic small molecules.”

“It is entirely possible that novel routes can be predicted for chemical warfare agents, circumventing national and international lists of watched or controlled precursor chemicals for known synthesis routes.”

The team from Collaborations Pharmaceuticals stressed its work was to raise awareness of the technology. “We should also remain diligent against the potential for dual-use, in the same way, that we would with physical resources such as molecules or biologics.”

Chemical weapons are at the forefront of the public consciousness at present given the ongoing war in Ukraine. The U.S. has said it will respond “aggressively” if Russia does deploy chemical weapons, according to various news reports.

Written by Ben Wodecki and republished with permission from AI Business.

This image has an empty alt attribute; its file name is aib_logo_1.png



Reach out to us at [email protected]

Related Posts

Procter & Gamble on Scaling AI for Enterprise

Procter & Gamble on Scaling AI for Enterprise

Data, talent, platforms and trust: these are the four key pillars firms need when building AI into their applications according…
Scaling Up: AWS’s Allie Miller Outlines the Key Trends in AI and Machine Learning

Scaling Up: AWS’s Allie Miller Outlines the Key Trends in AI and Machine Learning

We are entering an AI world where start-ups are transforming everyday lives through artificial intelligence and machine learning. AWS’ Allie…

ScaleUp:AI uses cookies to enhance your experience and help us analyze website usage. By continuing to browse or dismissing this banner, you indicate your agreement as outlined in our Privacy Policy.