The service is reportedly based on an algorithm that analyzes data about users of corporate devices and learns to distinguish them from each other. The program can detect signs of burnout in an employee, determine their status, and report if they are drunk or under the influence of drugs.
“During development, the team studied the effects of alcohol, fatigue, and other conditions on the characteristics of keyboard handwriting. For example, a bottle of beer with a strength of 4.5 degrees increases reaction time by 3-5% and the number of typos by 3-7%.” said.
It is noted that the service identifies the author of a particular text, even if he used a colleague’s computer.
Source: Ferra
