According to the patent, the algorithm would be able to look out for the user making typos, not precisely clicking on links and buttons, the way they are holding their phone, their walking speed and how long it takes to request a ride.The time of day, and where a ride is requested may also be considered.
At least that's according to a patent application spotted by CNN with the title "predicting user state using machine learning".
Uber patents tech that would identify drunk passengers before they get in the vehicle by analyzing a number of factors. Examples of interface interaction characteristics include the amount of time for a user to interact with the user interface after new information (or a modified display) is shown to the user, or the user's accuracy in pressing an interface element on the device.
However, a patent recently filed by the company reveals that new technology could enable Uber to identify intoxicated users and provide them with a reduced service - or no service at all.
Uber is all set to introduce the new app that will alert passengers whether the driver is drunk or not.
"It would be cool if drivers got extra money for picking up drunk passengers".
Ride-hailing services like Uber have transformed nightlife in car-centric cities, ditching the need for designated drivers and limiting intake of alcohol. Uber is said to be working on an Artificial Intelligence-powered system to pick out such inebriated passengers, and take steps accordingly.
Theoretically, knowing whether a passenger is intoxicated or exhausted would not result in the fare being denied, but would rather lead to the passenger being paired with a more experienced driver, Uber says.
The patent states that if a potential rider is acting uncharacteriscally, they could be rerouted to certain drivers who are experienced with handling troubled passengers, or not at all.