Embodiments of the disclosure relate to configuring a watermark unit with watermark algorithms for artificial intelligence (AI) models for a data processing (DP) accelerator. In one embodiment, in response to a request received by a DP accelerator, the request, sent by an application, to apply a watermark algorithm to an AI model by the DP accelerator, a system determines that the watermark algorithm is not available at a watermark unit of the DP accelerator. The system sends a request for the watermark algorithm. The system receives the watermark algorithm by the DP accelerator. The system configures the watermark unit at runtime with the watermark algorithm for the watermark algorithm to be used by the DP accelerator.

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.

In one embodiment, a computer-implemented method performed by a data processing (DP) accelerator, includes receiving, at the DP accelerator, first data representing a set of training data from a host processor; receiving, at the DP accelerator, a watermark kernel from the host processor; and executing the watermark kernel within the DP accelerator on an artificial intelligence (AI) model. The watermark kernel, when executed, is configured to: generate a new watermark by inheriting an existing watermark from a data object of the set of training data, train the AI model using the set of training data, and implant the new watermark within the AI model during training of the AI model. The DP accelerator then transmits second data representing the trained AI model having the new watermark implanted therein to the host processor.

Digital Witness is a solution based on advanced cryptographic techniques to ensure data integrity, authenticity, irrefutability and confidentiality at the point of data creation. The DigiWit process guarantee is based on using strong cryptographic techniques in conjunction with PKI and public/private block-chains. DigiWit process establishes a ‘root of trust’ for a digital artifact in conjunction with notarization provided by a trusted third-party. The result is a mathematical non-repudiable guarantee that the file under audit is exactly as recorded by the author. The authenticity of the author and the root of trust are provided by the notarizing trusted third-party. Integrity of the captured data is based on the time to insert its unique signature to the block-chain public ledger. This root of trust is intended to be permissible to prove authenticity of evidence in the legal arena (e.g., images of crime scenes, contracts, etc.) based on mathematical veracity.

A plug-in module, which, in combination with a host module, prevents unauthorized copying—like screen captures, screenshots, or screen recordings—of the streaming content provided to a participant in an online content-sharing session via an Instant Messenger (IM) service. The plug-in module may be a part of an IM application running on the participant's system or the host module may transmit a self-installing plug-in module to the participant's system upon receiving an indication that a user is hosting the online session. The plug-in module provides kernel-specific interface of the participant system's Operating System (OS) to the host module, which, then sends an OS-specific instruction to the plug-in module to trigger the OS to disable or control the copying of the streaming content as specified in a privacy preference received from the user hosting the online session. In this manner, sensitive and critical business data may be conveniently and securely shared online.

A method and computer software for creating an encoded image and which can optionally include a method for decoding the encoded image. The encoded image is preferably formed from at least one symmetric image but can be formed from a plurality of symmetric images. Embodiments of the present invention can be performed with physical paper and writing utensils or can be performed via computer software. Embodiments of the present invention can be used for art authentication based on results obtained by decoding an image. In one embodiment, one or more encoded image elements can be revealed simultaneously. Optionally, however, encoded image elements can be caused to be revealed in a series that gives a sense of motion in a manner similar to that of motion picture animation.

A system for facilitating a virtual screening is disclosed. The system identifies a user for participating in a virtual screening of media content. The system generates a unique link for the user to facilitate access to the media content. When the user interacts with the link, the system prompts the user to authenticate with the system and determines whether the authentication was successful. If authentication is successful, the link is associated with an identifier of the user and/or a user device. A request for a key for decrypting the media content and a request for a digitally signed file accessing the media content pursuant to parameters are made. If the digitally signed file is valid, the system applies a watermark to the media content to track the use of the media content and streams the media content to the user. Feedback on the media content is obtained from the user.

Methods, systems, and non-transitory computer readable storage media are disclosed for generating a machine-learning model and encoding ownership information in the machine-learning model. For example, the disclosed system can generate parameters of a machine-learning model utilizing digital content items modified by a filter. The disclosed system can then process digital content items modified by the filter to generate first outputs based on the digital content items being modified by the filter. The disclosed system can also process digital content items unmodified by the filter to generate second outputs based on the digital content items not being modified by the filter. The disclosed system can determine that the second outputs are degraded relative to the first outputs. Accordingly, the disclosed system can determine ownership of the machine-learning model based on detecting that information about the filter is embedded in parameters of the machine-learning model.

A video player for playing a video stream that receives a master playlist identifying at least one variant playlist identifying a video file encoded as a series of video frames that when decoded provide the video stream. The video player based upon a configuration tag in the master play list selectively determining whether the video file is to be processed in a trusted execution environment. The trusted execution environment of the video player selectively includes at least one of (i) hack one, only hack one; (ii) output and link protection; (iii) hardware root of trust; and (iv) forensic watermarking, and decrypts and/or decodes the video stream in such an environment.

A computer stores, within a single user account, multiple supervised computing resources and multiple additional computing resources. The multiple supervised computing resources are associated with a security policy. The computer executes a first instance of a specified application that lacks read access and lacks write access to any and all of the multiple supervised computing resources. The computer executes, simultaneously with the first instance, a second instance of the specified application that accesses at least a portion of the multiple supervised computing resources. The computer applies rules from the security policy to the second instance of the specified application while foregoing applying the rules from the security policy to the first instance of the specified application.