Looks like AI is going to revolutionize agriculture. New AI tools and advances in computing are providing groundbreaking insights into how plants interact with their environment. These breakthroughs could aid breeders in creating more resilient crops and help farmers prepare for future challenges in agriculture, Axios notes.

A growing global population and vulnerable plant life is behind this move. With the population booming, the pressure’s on to feed the masses with less land. But wait, there’s more: Issues with soil quality, bugs on a rampage, diseases on the loose, and climate change throwing a wild party, are also a factor.

This isn’t new. Robots are already capturing images of plants and using AI deep learning methods to ‘ detect diseases ’ and ‘ analyze plant attributes ’. This gives accurate and reliable data collection, in order for AI to play matchmaker — by analyzing plant traits, and predicting which genetic combos will churn out the superhero of crops a.k.a resistant to drought and disease. It's like Tinder for plants, only with fewer awkward dates and more bumper crops.

It’s also uncovering the complex details of plant biology that were once hidden. AlphaFold is one system that has increased understanding of plant proteins. Before AlphaFold, 2% of the proteins in Arabidopsis thaliana were known, but now we know about 76%.

Soil, climate and farming practices are also factors affecting plant growth. Studies show that machine learning can predict plant genetics based on soil microbiomes, suggesting potential for crops to support beneficial microbes and reduce chemical use.

But it’s not all rainbows and sunshine in the AI garden. Data availability and quality are major concerns, especially considering the scarcity of digitized images for analysis. There’s also a shortage of scientists skilled in both biology and proficient in computer science, and plant research isn’t trendy, in comparison to medical studies, which affects investment into the field.

Yet hope springs eternal in the world of AI all thanks to large language models (LLMs), which are being developed to translate the language of DNA and proteins. These tools hold promise in understanding how different genome regions influence plant traits, potentially reducing the need for extensive field testing.

Tags: