Lasers, robots and a lighter touch: AI learns to pull weeds, not drench fields

Weeds are farming's oldest chore, and the usual answers — spraying whole fields or back-breaking hand work — are getting harder. A new review from researchers at the University of Mississippi and Cornell University, published in the journal Engineering, looks at what happens when machines learn to spot a single weed in a row of crops: lasers that zap it, small robots that pull or till it out, and sprayers that put a drop exactly where it is needed and nowhere else. That precision can keep herbicide off the crop itself, and for organic farmers — who cannot use synthetic sprays and rely on scarce hand labour — it could be transformative. The reviewers are honest about steep costs and models trained on the wrong fields, but they call AI-guided weeding a possible 'third revolution' in how farms handle weeds, after synthetic herbicides and herbicide-resistant crops.

SDG 2 · Zero hunger — Mechanism: tech innovation