When a pharmaceutical company looked into whether its artificial-intelligence (AI) tools could be used to design biochemical weapons, the results horrified its researchers. Scientists used a machine-learning model that penalizes toxicity and inverted it to pursue compounds similar to the nerve agent VX, one of the most toxic chemical weapons ever created. In less than six hours, the system designed VX and many other known chemical-warfare agents, as well as molecules predicted to be even more toxic. The frightening ease of the experiment should be a wake-up call for the AI drug-discovery community, argued Fabio Urbina and three colleagues in March. “By going as close as we dared, we have still crossed a grey moral boundary,” they write. “We can easily erase the thousands of molecules we created, but we cannot delete the knowledge of how to recreate them.”
Read the full text here.
References
- Urbina, F. L., Lentzos, F., Invernizzi, C. F., & Ekins, S. (2022). Dual use of artificial-intelligence-powered drug discovery. Nature Machine Intelligence, 4(3), 189–191. https://doi.org/10.1038/s42256-022-00465-9
Disclaimers
- The material in these reviews is from various public open-access sources, meant for educational and informational purposes only
- Any personal opinions expressed are those of only the author(s) and are not intended to represent the position of any organization(s)
- No official support by any organization(s) has been provided or should be inferred
