Granted Patent B2

US 12,493,775 · App. 19/036,141 · Granted Dec 9, 2025

Inventors: Soumitri Kolavennu (Minneapolis, MN); Raj Bharadwaj (Minneapolis, MN); Lavanya Basavaraju (Minneapolis, MN); Jayant Chawla (Minneapolis, MN); Christopher Ziolkowski (Burnsville, MN); Valerie Lancelle (Caledonia, WI); Christopher Davis (Minneapolis, MN)

Assignee: U.S. BANK NATIONAL ASSOCIATION

G06N3/0475G06F11/1446G06F16/90344G06F30/27G06F16/2462G06F17/18G06F18/22G06N20/00

Abstract

Claims (30)

1 . A method comprising:

obtaining initial content;

generating a prompt based on the initial content;

selecting a set of generators;

for each respective generator of the set of generators:

using the prompt to produce new content;

comparing the new content to the initial content; and

making a respective determination whether the new content and the initial content are similar based on whether a similarity threshold is satisfied;

using the respective determinations of the set of generators, ensuring that an authorship token is associated with the initial content,

wherein the authorship token includes a form indicating that an associated content region has human rather than an artificial authorship or artificial rather than human authorship.

2 . The method of claim 1 , further comprising;

determining that similar content to the initial content is not present in the data store,

wherein the generating of the prompt is responsive to determining that the similar content is not present in the data store.

3 . The method of claim 1 , wherein generating the prompt based on the initial content includes:

providing at least a portion of the initial content as input into a large language model that is trained or prompted to produce output comprising the prompt based on the input.

4 . The method of claim 1 , wherein generating the prompt includes:

extracting features, key words, themes, or other characteristics of the initial content; and

using a template to put such characteristics into a prompt that includes an outline or process for generating content.

5 . The method of claim 1 , wherein the prompt is configured to mimic what a user that wanted to use artificial intelligence to generate content would have provided to an artificial intelligence.

6 . The method of claim 1 , wherein selecting the set of generators includes:

selecting the set of generators based on a content type of the initial content; and

selecting the set of generators based on a likelihood that respective generators were used to produce the initial content.

7 . The method of claim 1 , wherein whether a similarity threshold is satisfied includes:

determining a distance between at least a portion of the initial content and the new content, wherein the distance is determined based on edit distance or distance in embedding space.

8 . The method of claim 1 , further comprising:

determining that the initial content is generic and based thereon determining that the initial content either lacks artificial authorship or has default authorship.

9 . The method of claim 1 , further comprising:

sending a communication to a server or calling an application programming interface associated with a server regarding at least one authorship token.

10 . The method of claim 1 , further comprising:

storing at least one of the authorship tokens in a blockchain.

Assignments (1)

Continuity (7)

References Cited (93)