The technology described herein obfuscates image content using a local neural network and a remote neural network. The local network runs on a local computer system and a remote classifier runs in a remote computing system. Together, the local network and the remote classifier are able to classify images, while the image never leaves the local computer system. In aspects of the technology, the local network receives a local image and creates a transformed object. The transformed object may be generated by processing the image with a local neural network to generate a multidimensional array and then randomly shuffling data locations within a multidimensional array. The transformed object is communicated to the remote classifier in the remote computing system for classification. The remote classifier may not have the seed used to deterministically scramble the spatial arrangement of data within the multidimensional array.

Wireless communication methods are described for a user plane integrity protection failure detection and handling, determination and management of integrity protection enabled data rate that exceeds or is close to exceeding a user equipment's capability or threshold, and management of integrity protection or encryption mechanisms in a dual-connectivity system that includes a master network node and a secondary network node.

Described are methods, systems and computer readable media for data source refreshing.

In one embodiment, a method for automatic inference of meeting attendance is provided. The method comprises sending a calendar request to a plurality of users that are invited to a meeting. The method further comprises receiving from each user of the plurality, a unique string identifying the user. The method further comprises generating a lookup table identifying the users of the plurality and their respective unique strings. The method further comprises receiving a first string broadcasted by a first user during the meeting. The method further comprises, responsive to determining that the broadcasted first string does not match one of the unique strings in the lookup table, performing an action to prevent the first user from receiving meeting content determined to be confidential.

Systems, devices, and techniques are generally described to transmit image data. An image file including image data may be identified. The image data may be divided into a plurality of portions. The plurality of portions may include a first portion corresponding to a first position in the image data and a second portion corresponding to a second position in the image data. The image data may be modified to produce a shuffled image. The shuffled image may indicate that the first portion corresponds to a third position and the second portion corresponds to a fourth position. The shuffled image may be sent to a recipient computing device. The recipient computing device may render the first portion at the first position and the second portion at the second position to display the image data.

An image processing system includes circuitry. The circuitry accepts, from a user, an input of information to be registered. The circuitry registers the inputted information as registered information. The circuitry generates tracing data to be used for tracing the registered information. The circuitry outputs the tracing data. The circuitry associates access authorization to an access log for the registered information with the tracing data. The circuitry accepts an input of the tracing data. The circuitry acquires the access log that is associated with the inputted tracing data, and displays the acquired access log.

The present technology relates to identifying an artificial intelligence model based on a received first key value to write a received first block of data associated with a file. The received first key value is applied to the identified artificial intelligence model which is trained to output one of a plurality of actual index values where the identified artificial intelligence model and the plurality of data blocks are stored as a neural tree. The one of the actual index values is compared to a range within the actual index values to determine when the one of the actual index value points to a first data block of the plurality of data. The received first block of data associated with the file is written into the determined first data block.

A method for secure access control to a data storage system for a host apparatus by means of an access control device, the method comprising: as part of a first mode of operation of the access control device, receiving user data from the host apparatus and transmitting it in unmodified or modified form to the data storage system for local storage; exchanging a first cryptographic secret with a computer system to enable encryption of data by the access control device in dependence on the first cryptographic secret; receiving a data read request for at least a portion of the user data stored in the data storage system; in response to the data read request, transitioning the access control device to a second mode of operation in which the access control device is configured to perform read access but not write or delete access to the data storage system; and in the second operating mode, retrieving user data requested according to the data read request from the data storage system, encrypting them using the first cryptographic secret (K) or a key derived therefrom according to a key generation rule and transmitting the user data encrypted in this way to a predetermined user data recipient; wherein the user data is processed as part of the method in such a way that the encrypted user data transmitted as part of the second operating mode represents information which can be extracted from it for the user data recipient and which represents an identity of the access control device and/or of the data storage system or allows a clear conclusion to be drawn therefrom.

To train models, training data is needed. As personal data changes over time, the training data can get stale, obviating its usefulness in training the model. Embodiments deal with this by developing a database with a running log specifying how each person's data changes at the time. When data is ingested, it may not he normalized. To deal with this, embodiments clean the data to ensure the ingested data fields are normalized. Finally, the various tasks needed to train the model and solve for accuracy of personal data can quickly become cumbersome to a computing device. They can conflict with one another and compete inefficiently for computing resources, such as processor power and memory capacity. To deal with these issues, a scheduler is employed to queue the various tasks involved.

An information provision apparatus includes a memory configured to store personal data for each user, and a processor coupled to the memory and configured to in response to receiving a request for first personal data of a first user from a terminal device, determine difference between first data stored in the memory as the first personal data at a first time of receiving the request and second data stored in the memory as the first personal data at a second time before the first time, provision of the second data being permitted, perform, in accordance with the difference, determination of whether provision of the first data is permitted, and when it is determined that the provision of the first data is permitted, transmit the first data to the terminal device.