Abstract: Disclosed are systems and methods for uniquely identifying a hardware device. In one aspect, a method may comprise (a) obtaining a first partial key and encrypted parameters from a database and a second partial key from a remote server; (b) decrypting the encrypted parameters using the first partial key and the second partial key, to thereby generate decrypted parameters; (c) obtaining attributes of a hardware device, wherein the attributes comprise a state of a CPU or GPU of the hardware device; (d) processing, on the hardware device, the attributes with a first ML algorithm to generate a digital fingerprint of the hardware device, wherein the first ML algorithm comprises the decrypted parameters; and (e) processing, on the remote server, at least the digital fingerprint of the hardware device and the attributes with a second ML algorithm configured to determine whether the hardware device has been previously identified.

Abstract: A machine detects a request to execute a transaction specified by a first set of user inputs. The machine determines, based on the first set of user inputs that specified the transaction, that the request to execute the transaction is to be verified with a corresponding challenge prompt that is to be generated for the request to execute the transaction. The machine then generates the challenge prompt that corresponds to the request to execute the transaction specified by the first set of user inputs that specified the transaction, and the machine causes presentation of the generated challenge prompt that corresponds to the request to execute the transaction. In response to the presented challenge prompt, the machine may receive a second set of user inputs. Based on the second set of user inputs, the machine then generates an indication of whether the request is verified.

Abstract: A machine trains an artificial intelligence engine to facilitate authentication of a request to verify a user. The machine accesses a reference set of obfuscated geolocations generated from actual geolocations from which a device submitted requests to verify the user. The machine groups the obfuscated geolocations into geographical clusters based on a predetermined cluster radius value and calculates a corresponding representative geolocation for each geographical cluster and a corresponding variance distance from the representative geolocation for each geographical cluster. The machine then generates a reference location score based on the representative geolocations of the geographical clusters and on the variance distances of the geographical clusters. The machine trains an artificial intelligence engine to output that reference location score in response to the reference set being input thereto. The trained artificial intelligence engine may then be provided to one or more devices.

Abstract: A machine detects a request to execute a transaction specified by a first set of user inputs. The machine determines, based on the first set of user inputs that specified the transaction, that the request to execute the transaction is to be verified with a corresponding challenge prompt that is to be generated for the request to execute the transaction. The machine then generates the challenge prompt that corresponds to the request to execute the transaction specified by the first set of user inputs that specified the transaction, and the machine causes presentation of the generated challenge prompt that corresponds to the request to execute the transaction. In response to the presented challenge prompt, the machine may receive a second set of user inputs. Based on the second set of user inputs, the machine then generates an indication of whether the request is verified.

Abstract: A device authenticates a request to verify a user. The device accesses a face image that depicts a face of the person and includes a characteristic noise pattern inserted by a camera of the device. The device also accesses a geolocation at which the device captured the face image and inputs the face image and the geolocation into an artificial intelligence engine that outputs a face score, a device score, and a location score. The device next submits the request with the scores to a server machine and obtains an authentication score from the server machine. The device then presents an indication that the request to verify the person is authentic based on a comparison of the obtained authentication score to a threshold authentication score.

Abstract: Disclosed are systems, methods, and non-transitory computer-readable media for authentication based on physical interaction and characteristic noise patterns. Execution of a requested transaction may be conditioned upon satisfaction of an authentication requirement. For example, the requesting user may be prompted to perform a physical interaction such as a swipe across a screen of a client device. The sensor data includes a characteristic noise pattern caused by manufacturing deviations of the set of sensors that captured the sensor data. The sensor data describing the physical interaction and the characteristic noise pattern are used to determine whether the authentication requirement has been satisfied. For example, the sensor data and characteristic noise pattern are used to determine whether the user that performed the physical interaction is an authorized user. The authentication requirement is satisfied upon determining that the user that performed the physical interaction is an authorized user.

Abstract: Disclosed are systems, methods, and non-transitory computer-readable media for detecting fraudulent transactions. A fraud detection system determines whether transactions are fraudulent based on a machine learning model framework that leverages a combination of transaction data describing a transaction and sequence data describing a sequence of event preceding the transaction. The machine learning model framework includes multiple feature models that each provide an output based on a different set of feature data describing the transaction, as well an events sequence model that provides an output based on the sequence data describing the sequence of event preceding the transaction. The output of these machine learning models is used to generate a cumulative input that is provided into a secondary machine learning model that outputs a probability value indicating a likelihood that the transaction is fraudulent.

Abstract: A device authenticates a request to verify a user. The device accesses a face image that depicts a face of the person and includes a characteristic noise pattern inserted by a camera of the device. The device also accesses a geolocation at which the device captured the face image and inputs the face image and the geolocation into an artificial intelligence engine that outputs a face score, a device score, and a location score. The device next submits the request with the scores to a server machine and obtains an authentication score from the server machine. The device then presents an indication that the request to verify the person is authentic based on a comparison of the obtained authentication score to a threshold authentication score.

Abstract: A device authenticates a request to verify a user. The device accesses a face image that depicts a face of the person and includes a characteristic noise pattern inserted by a camera of the device. The device also accesses a geolocation at which the device captured the face image and inputs the face image and the geolocation into an artificial intelligence engine that outputs a face score, a device score, and a location score. The device next submits the request with the scores to a server machine and obtains an authentication score from the server machine. The device then presents an indication that the request to verify the person is authentic based on a comparison of the obtained authentication score to a threshold authentication score.

Abstract: A machine trains an artificial intelligence engine to facilitate authentication of a request to verify a user. The machine accesses a reference set of obfuscated geolocations generated from actual geolocations from which a device submitted requests to verify the user. The machine groups the obfuscated geolocations into geographical clusters based on a predetermined cluster radius value and calculates a corresponding representative geolocation for each geographical cluster and a corresponding variance distance from the representative geolocation for each geographical cluster. The machine then generates a reference location score based on the representative geolocations of the geographical clusters and on the variance distances of the geographical clusters. The machine trains an artificial intelligence engine to output that reference location score in response to the reference set being input thereto. The trained artificial intelligence engine may then be provided to one or more devices.

Abstract: A machine trains an artificial intelligence engine to facilitate authentication of a request to verify a user. The machine accesses a reference set of obfuscated geolocations generated from actual geolocations from which a device submitted requests to verify the user. The machine groups the obfuscated geolocations into geographical clusters based on a predetermined cluster radius value and calculates a corresponding representative geolocation for each geographical cluster and a corresponding variance distance from the representative geolocation for each geographical cluster. The machine then generates a reference location score based on the representative geolocations of the geographical clusters and on the variance distances of the geographical clusters. The machine trains an artificial intelligence engine to output that reference location score in response to the reference set being input thereto. The trained artificial intelligence engine may then be provided to one or more devices.

Abstract: A device authenticates a request to verify a user. The device accesses a face image that depicts a face of the person and includes a characteristic noise pattern inserted by a camera of the device. The device also accesses a geolocation at which the device captured the face image and inputs the face image and the geolocation into an artificial intelligence engine that outputs a face score, a device score, and a location score. The device next submits the request with the scores to a server machine and obtains an authentication score from the server machine. The device then presents an indication that the request to verify the person is authentic based on a comparison of the obtained authentication score to a threshold authentication score.

Abstract: A device authenticates a request to user. The device accesses a face image that depicts a face of the person and includes a characteristic noise pattern inserted by a camera of the device. The device also accesses a geolocation which the device captured the face image and puts the face image and the geolocation into an artificial intelligence engine that outputs a face score, device score, and a location score. The device next submits the request with the scores to a server machine and obtains an authentication score from the server machine. The device then presents an indication that the request to verify the person is authentic based on a comparison of the obtained authentication score to a threshold authentication score.

Abstract: A device authenticates a request to user. The device accesses a face image that depicts a face of the person and includes a characteristic noise pattern inserted by a camera of the device. The device also accesses a geolocation which the device captured the face image and puts the face image and the geolocation into an artificial intelligence engine that outputs a face score, device score, and a location score. The device next submits the request with the scores to a server machine and obtains an authentication score from the server machine. The device then presents an indication that the request to verify the person is authentic based on a comparison of the obtained authentication score to a threshold authentication score.

Abstract: A machine trains an artificial intelligence engine to facilitate authentication of a request to verify a user. The machine accesses a reference set of obfuscated geolocations generated from actual geolocations from which a device submitted requests to verify the user. The machine groups the obfuscated geolocations into geographical clusters based on a predetermined cluster radius value and calculates a corresponding representative geolocation for each geographical cluster and a corresponding variance distance from the representative geolocation for each geographical cluster. The machine then generates a reference location score based on the representative geolocations of the geographical clusters and on the variance distances of the geographical clusters. The machine trains an artificial intelligence engine to output that reference location score in response to the reference set being input thereto. The trained artificial intelligence engine may then be provided to one or more devices.