In 1989, computer scientist Yann LeCun and colleagues focused on the automated reading of messy, handwritten zip codes on mail envelopes for the US Postal Service. The problem became an ultimate test for computer vision, showing that machines can distinguish complex visual patterns without human assistance. Prior to this, computers used rigid algorithms where a programmer would instruct the computer to look for a number seven in its form of one horizontal line and one slanted vertical line, but human handwriting often did not match this mathematical definition. As stated by a 2024 publication in Cancers, the 1989 project represented the point where LeCun and colleagues used a convolutional neural network to read handwritten digits, analyzing small parts of each image and combining visual clues through layers to recognize variations in handwriting. This was not just a laboratory experiment but a solution for sorting thousands of envelopes per hour.