Scientists built a type of bacteria that can actually 'learn' how to play Tic-Tac-Toe by saving its memories in its own DNA.
March 20, 2026
Original Paper
A genetically encoded local learning rule enables physical learning in engineered bacteria
bioRxiv · 10.64898/2026.03.18.712691
The Takeaway
Researchers used gene-edited E. coli to create a biological version of a neural network that stores 'weights' as DNA copy counts. They successfully trained these bacterial colonies to compete in a game tournament, proving that living matter can perform complex learning tasks usually reserved for AI.
From the abstract
Training physical neural networks directly in matter remains difficult because most platforms do not implement weight storage and weight update within the same physical substrate. Here we show that engineered Escherichia coli can implement a genetically encoded local learning rule acting on a persistent biological memory. In memregulons, analogue weights are stored as plasmid copy-number ratios in a coupled two-plasmid system and are rewritten by activity-dependent growth bias under a global neg