Back to results
US 12,608,178 B2Granted
Auditable authorship attribution with automatically applied authorship tokens
AssigneeU.S. Bank National Association
InventorsChristopher Ziolkowski, Valerie Lancelle, Christopher Davis
Life of this patent
Estimating the term…
Abstract
A content editor or a plugin thereto automatically generates authorship tokens that identify content authored by a human author or an artificial author. The authorship tokens are applied to the work while the work is being produced. Thus, subsequent review of the work can identify regions produced by a human author and other regions produced by an artificial intelligence.
Claims20
A computer-implemented method for improving word processing software, the method comprising:
- providing, via a word processor running on a computing environment, a first user interface element actuatable to cause activation of an authorship tracking feature of the word processor;
- based on the authorship tracking feature being activated and automatically during one or more edit sessions where content is being edited using the word processor associated with a computing environment:
- receiving, over a keyboard human interface device communicatively coupled to the computing environment, a first edit comprising one or more characters to insert into a first region of the content using the word processor and at a location specified by a first cursor location in the content;
- inserting, by the word processor and within the first region of the content at the location specified by the cursor location, human text input entered via the human interface device;
- determining that a human rather than an artificial intelligence made the first edit based on the first edit being caused by input received over the keyboard human interface device of the computing environment associated with the word processor;
- determining that the first edit satisfies a threshold of significance;
- identifying the first region associated with the first edit, the first region being selected from the group consisting of:
- word level, character level, sentence level, paragraph level, and a file level;
- and
- responsive to determining the first region associated with the first edit satisfies the threshold of significance and responsive to determining that the human made the first edit rather than an artificial intelligence, providing a first authorship token in association with the first region, wherein the first authorship token has a predetermined form indicating that a human rather than an artificial intelligence is the author of the content in the first region;
- determining that a human rather than an artificial intelligence made a second edit to a second region of the content at a second cursor location using the word processor;
- associating the second region with a second token indicating that the second region has human rather than artificial authorship;
- after associating the second region with the second token, detecting activation of a second user interface element;
- responsive to detecting the activation of the second user interface element, causing the word processor to activate a generative artificial intelligence function that causes an artificial edit to modify the second region;
- determining that the artificial edit is from an artificial source;
- responsive to the determining that the artificial edit is from an artificial source, associating the second region with a third authorship token having a form indicating that the second region has mixed human authorship and artificial authorship;
- inserting import content into a third region of the content, the import content being from outside of the word processor and inserted into the third region as a result of a paste operation, a drag-and-drop operation, or selection of an import operation from a menu of the word processor;
- associating the third region with a fourth authorship token that indicates the third region is associated with external content;
- and
- generating a report showing authorship statistics for the content based at least on the first, third, and fourth authorship tokens, wherein the authorship statistics comprise an indication of a metadata that describes a number of or percentage of characters, words, or other delimitations of content that have certain kinds of authorship,
- wherein determining that the first edit satisfies the threshold of significance includes accounting for an output from performing of one or more operations selected from the group consisting of:
- determining whether the first edit includes pasting;
- determining whether the first edit includes moving the content;
- determining an amount of AI or artificial assistance used;
- determining a level of sophistication of a word processor feature that was used to produce the first edit;
- determining that the edit changes more than a predetermined threshold percentage of characters or words within the first region;
- and
- determining that the edit changes more than a predetermined threshold number of characters or words.
25 more elements
A non-transitory computer-readable storage medium having computer-executable instructions thereon which, when executed by a computing system, perform a method for improving word processing software, the method comprising:
- providing, via a word processor running on a computing environment, a first user interface element actuatable to cause activation of an authorship tracking feature of the word processor;
- based on the authorship tracking feature being activated and automatically during one or more edit sessions where content is being edited using the word processor associated with a computing environment:
- receiving, over a keyboard human interface device communicatively coupled to the computing environment, a first edit comprising one or more characters to insert into a first region of the content using the word processor and at a location specified by a first cursor location in the content;
- inserting, by the word processor and within the first region of the content at the location specified by the cursor location, human text input entered via the human interface device;
- determining that a human rather than an artificial intelligence made the first edit based on the first edit being caused by input received over the keyboard human interface device of the computing environment associated with the word processor;
- determining that the first edit satisfies a threshold of significance;
- identifying the first region associated with the first edit, the first region being selected from the group consisting of:
- word level, character level, sentence level, paragraph level, and a file level;
- and
- responsive to determining the first region associated with the first edit satisfies the threshold of significance and responsive to determining that the human made the first edit rather than an artificial intelligence, providing a first authorship token in association with the first region, wherein the first authorship token has a predetermined form indicating that a human rather than an artificial intelligence is the author of the content in the first region;
- determining that a human rather than an artificial intelligence made a second edit to a second region of the content at a second cursor location using the word processor;
- associating the second region with a second token indicating that the second region has human rather than artificial authorship;
- after associating the second region with the second token, detecting activation of a second user interface element;
- responsive to detecting the activation of the second user interface element, causing the word processor to activate a generative artificial intelligence function that causes an artificial edit to modify the second region;
- determining that the artificial edit is from an artificial source;
- responsive to the determining that the artificial edit is from an artificial source, associating the second region with a third authorship token having a form indicating that the second region has mixed human authorship and artificial authorship;
- inserting import content into a third region of the content, the import content being from outside of the word processor and inserted into the third region as a result of a paste operation, a drag-and-drop operation, or selection of an import operation from a menu of the word processor;
- associating the third region with a fourth authorship token that indicates the third region is associated with external content;
- and
- generating a report showing authorship statistics for the content based at least on the first, third, and fourth authorship tokens, wherein the authorship statistics comprise an indication of a metadata that describes a number of or percentage of characters, words, or other delimitations of content that have certain kinds of authorship,
- wherein determining that the first edit satisfies the threshold of significance includes accounting for an output from performing of one or more operations selected from the group consisting of:
- determining whether the first edit includes pasting;
- determining whether the first edit includes moving the content;
- determining an amount of AI or artificial assistance used;
- determining a level of sophistication of a word processor feature that was used to produce the first edit;
- determining that the edit changes more than a predetermined threshold percentage of characters or words within the first region;
- and
- determining that the edit changes more than a predetermined threshold number of characters or words.
25 more elements
A computing system comprising:
- a processor;
- a non-transitory memory communicatively connected to the processor and storing instructions which, when executed by the computing system, cause the computing system to:
- provide, via a word processor running on a computing environment, a first user interface element actuatable to cause activation of an authorship tracking feature of the word processor;
- based on the authorship tracking feature being activated and automatically during one or more edit sessions where content is being edited using the word processor associated with a computing environment:
- receive, over a keyboard human interface device communicatively coupled to the computing environment, a first edit comprising one or more characters to insert into a first region of the content using the word processor and at a location specified by a first cursor location in the content;
- insert, by the word processor and within the first region of the content at the location specified by the cursor location, human text input entered via the human interface device;
- determine that a human rather than an artificial intelligence made the first edit based on the first edit being caused by input received over the keyboard human interface device of the computing environment associated with the word processor;
- determine that the first edit satisfies a threshold of significance;
- identify the first region associated with the first edit, the first region being selected from the group consisting of:
- word level, character level, sentence level, paragraph level, and a file level;
- and
- responsive to determining the first region associated with the first edit satisfies the threshold of significance and responsive to determining that the human made the first edit rather than an artificial intelligence, provide a first authorship token in association with the first region, wherein the first authorship token has a predetermined form indicating that a human rather than an artificial intelligence is the author of the content in the first region;
- determine that a human rather than an artificial intelligence made a second edit to a second region of the content at a second cursor location using the word processor;
- associate the second region with a second token indicating that the second region has human rather than artificial authorship;
- after associating the second region with the second token, detect activation of a second user interface element;
- responsive to detecting the activation of the second user interface element, cause the word processor to activate a generative artificial intelligence function that causes an artificial edit to modify the second region;
- determine that the artificial edit is from an artificial source;
- responsive to the determining that the artificial edit is from an artificial source, associate the second region with a third authorship token having a form indicating that the second region has mixed human authorship and artificial authorship;
- insert import content into a third region of the content, the import content being from outside of the word processor and inserted into the third region as a result of a paste operation, a drag-and-drop operation, or selection of an import operation from a menu of the word processor;
- associate the third region with a fourth authorship token that indicates the third region is associated with external content;
- and
- generate a report showing authorship statistics for the content based at least on the first, third, and fourth authorship tokens, wherein the authorship statistics comprise an indication of a metadata that describes a number of or percentage of characters, words, or other delimitations of content that have certain kinds of authorship,
- wherein determining that the first edit satisfies the threshold of significance includes accounting for an output from performing of one or more operations selected from the group consisting of:
- determining whether the first edit includes pasting;
- determining whether the first edit includes moving the content;
- determining an amount of AI or artificial assistance used;
- determining a level of sophistication of a word processor feature that was used to produce the first edit;
- determining that the edit changes more than a predetermined threshold percentage of characters or words within the first region;
- and
- determining that the edit changes more than a predetermined threshold number of characters or words.
27 more elements
Dependencies are inferred from the claim text.