GEN DIGITAL INC.

Obfuscation techniques used in phishing attacks that evade detection using current machine learning models can be identified and statistically characterized for use in generating adversarial training samples to improve the performance of a machine learning model phishing classifier. By using statistical information regarding the prevalence and effectiveness of various obfuscation techniques observed in real-world samples, adversarial samples that are effective and realistic examples of phishing content that evades phishing content classifier detection can be generated and used to augment the data set used for training or re-training the machine learning model classifier.

A method for accessing a network resource including detecting an attempt by a user via a computing device to access a service enabled by a computing system via a network and transmitting via the network to the computing system a first request to access the service in response to detecting the attempt by the user to access the service, the first request including at least one empty personally identifiable data structure. A failure to access the service responsive to the first request is determined. A second request to access the service in response to the first failure to access the service is transmitted via the network to the computing system, the second request including artificial personally identifiable information, and access to the service from the computing system is received for the user.

The disclosed computer-implemented method for detecting user incapacitated states for securing a mobile device against unintended use may include (i) monitoring sensor data generated for a current user of the mobile device, (ii) detecting, based on the sensor data, an incapacitated state associated with the current user, and (iii) performing a security action that protects against unintended use of the mobile device by disabling access to device functionality when the current user is in the incapacitated state. Various other methods, systems, and computer-readable media are also disclosed.

Systems and methods for transacting over a network. A first agent and a second agent are provided. The second agent is operable to transact with a third agent for use of a network-enabled service based on a first transaction policy from a fourth agent, the third agent enabled to communicate with a fifth agent. The first agent is operable to communicate with the second agent to facilitate the transacting by the second agent with the third agent for the use of the network-enabled service based on the first transaction policy and communicate with the fifth agent to facilitate the transacting by the second agent with the third agent for the use of the network-enabled service.

A method and system detects at a plurality of network locations a plurality of accuracy ratings of a plurality of media instances and detects the plurality of media instances. A particular accuracy rating of one or more particular media instances is detected at a particular network location, and the one or more particular media instances are detected. A bias of the particular accuracy rating is determined based on the particular accuracy rating, the one or more particular media instances, the plurality of accuracy ratings, and the plurality of media instances. An indication is transmitted to a user based on the bias of the particular accuracy rating.

A malware classification system provides improved confidence in explanations of neural network classification outputs using methods such as weighting or masking when training the neural network to train the network on a sample resembling or including the explanation. The explanation in some examples comprises a subset of a hierarchical input vector that is responsible for the neural network's classification output. In another example the neural network has an inner portion configured to reduce the weight of elements of the output not significantly contributing to the explanation of the output, such as by reducing the weight of as many such outputs to zero as is practical in generating the desired output.

Systems and methods for in-transit email review and data collection. In one embodiment, a method may include receiving a plurality of original emails at an alias email address. The alias email address may be associated with a user. Once a plurality of original emails are received at the alias email address, the method may include analyzing the plurality of original emails to identify relevant data contained within the plurality of original emails, and storing a copy of the identified relevant data in a database that is accessible by the user.

Systems and methods for classifying user intent based on the user interaction with an entry bar on a user device include receiving an input from a user in an entry box; classifying the input into a category of a plurality of categories including a category for an Artificial Intelligence (AI) session for an AI agent; and performing an action responsive to the classified category, including utilizing the AI agent for the AI session when the category is the category for the AI session and bypassing the AI agent for other categories of the plurality of categories.

Systems and methods for transacting over a network is provided. The system includes a first agent and second agent. The first agent is operable to receive from a third agent a transaction code associated with one or more credential types required to apply the transaction code or with one or more credential claim types required to apply the transaction code, transmit the transaction code to the second agent, and receive from the second agent a digitally signed transaction, a first verifiable proof, and the transaction code. The first agent is further operable to transmit to a fourth agent a second verifiable proof based on the first verifiable proof and the transaction code, receive from the fourth agent an unlock signature for a locked credential including one or more credential claims, and transmit the unlock signature to the second agent.

A system and method are provided by which an electronic address associated with a user is monitored. Based on the monitoring, an electronic message is detected including a digital document. A cryptographic function is applied to the digital document to generate a hash which is rendered accessible at a network location. An identification of the network location of the hash is transmitted to a first computing system associated with the user.