Researchers train a special-purpose deep-learning algorithm that can guess 4-digit card PINs 41% of the time, even if the victim is covering the pad with their hands. The machine-learning model is trained to recognize pad presses and assign specific probabilities on a set of guesses, using video of people typing PINs on the ATM pad. Concealing a pinhole camera at the top of the ATM was determined to be the best approach for the attacker. The researchers used the experiment’s video clips on a survey with 78 participants to determine if humans could also guess the concealed PINs.”]