MIT will use generative AI to design safer herbicide candidates
The MIT-led PLOW team will use generative AI to design new herbicide molecules and RNAi candidates intended to reduce risks to people and ecosystems.
AT A GLANCE
Key points
- PLOW will use generative AI to design chemical and RNAi herbicide candidates.
- The models will optimize cost alongside resistance, health, and ecological risks.
- AI-generated candidates will be validated through screening and greenhouse studies.
What is changing?
An MIT-led team has announced the PLOW project, which can receive up to $53 million through ARPA-H’s STREAM program. The project will use generative AI to predict chemical molecules or small interfering RNA (siRNA) candidates that could disrupt the same plant enzymes targeted by glyphosate and other herbicides. The models will optimize several constraints together, including synthesis cost, the risk of resistance, and potential harm to people and ecosystems.
According to MIT News, the models will generate billions of candidates; a high-throughput screening team will test this pool and feed results back into the computational system. Shortlisted candidates will move into greenhouse studies, with the first RNAi candidates planned for greenhouse testing by the end of the program’s first year. The team will also develop living microbial systems that can break down agricultural chemicals. The work remains at the research stage, and outcomes depend on laboratory and greenhouse validation.
Explore the original sources
Source date: 10 November 2026.
Published by AI Radar: 10 Nov 2026, 09:08 (Istanbul, UTC+3).
FOLLOW THE TOPIC
Topics and tags
Share your discovery.
Send this story to a friend.
Reader comments
What do you think about this development? Share your experience or question. Comments are published after editorial approval.
Your name/nickname and comment become public if approved. Read: Privacy notice.
Loading comments…