Automatic data access from derived trust level

A system for automatically granting access to user data comprising a hardware processor and computer executable instructions, that when executed by the hardware processor, cause the system to perform operations comprising:

• receiving a plurality of interactions from a first user to a second user, each interaction comprising an identity of the second user, information communicated to the second user, and a context in which the information was communicated;
• deriving a trust model from the plurality of interactions using a machine learning technique;
• without user intervention, annotating a larger body of first user data, the annotation indicating what trust level or security model setting grants access to the annotated first user data;
• receiving a communication from the first user comprising:
• a user identity;
• identification of first user data communicated to a user associated with the user identity, the first user data being a part of the larger body of first user data;
• and
• a context in which the communication is made;
• selecting the trust model from among a plurality of trust models based on the identity of the user;
• identifying a subset of the larger body of first user data based on the trust model that the user is allowed access to;
• identifying a portion of the subset that is relevant to the communication;
• and
• injecting the portion into a conversation between the first user and the user.

A computer storage medium comprising executable instructions that, when executed by a processor of a machine, cause the machine to perform operations comprising:

• receiving a plurality of interactions from a first user to a second user, each interaction comprising an identity of the second user, information communicated to the second user, and a context in which the information was communicated;
• deriving a trust model from the plurality of interactions using a machine learning technique;
• deriving a trust level for each item in a body of first user data using the trust model;
• without user intervention, annotating each item with a corresponding trust level, or adjusting a security model setting required to access each item, or both;
• receiving information comprising a request for the first user data or a communication from the first user, the information comprising:
• a user identity;
• identification of the first user data;
• and
• a context in which the request or communication is made;
• selecting the trust model from among a plurality of trust models based on the user identity;
• presenting the request or communication to the trust model and in response to the request or communication, receiving a trust level or a security model setting;
• and
• granting or denying access to the first user data based on the trust level or the security model setting.