MIT leads $53 million project to develop AI-driven weed killers

According to MIT Biological Engineering:

Project Background

Modern agriculture relies heavily on herbicides like glyphosate, but years of use have led to resistance mutations in weeds. At the same time, concerns are growing over the potential health and environmental risks associated with these chemicals. To address this, the U.S. Advanced Research Projects Agency for Health (ARPA-H) has awarded up to $53 million to an MIT-led team for the PLOW project.

AI and Research Strategy

The team will develop generative AI models to predict new chemical compounds or small interfering RNA (siRNA) that disrupt the same plant enzymes as traditional herbicides without triggering resistance. These models will optimize for low cost, safety, and effectiveness.

Testing and Implementation

The generated candidates will undergo high-throughput screening by partner institutions. A third team will then conduct greenhouse studies, with the first tests expected within a year. Additionally, the project will develop engineered microbes to safely degrade agrochemicals before they reach the environment.