US 12,393,577 B2
Techniques and architecture for securing large language model assisted interactions with a data catalog
David Hermann Peter Weik, Viernheim (DE)
Assigned to SAS INSTITUTE INC., Cary, NC (US)
Filed by SAS INSTITUTE INC., Cary, NC (US)
Filed on Oct. 2, 2024, as Appl. No. 18/904,206.
Claims priority of provisional application 63/607,982, filed on Dec. 8, 2023.
Claims priority of provisional application 63/545,729, filed on Oct. 25, 2023.
Prior Publication US 2025/0139088 A1, May 1, 2025
Int. Cl. G06F 16/242 (2019.01); H04L 9/40 (2022.01)
CPC G06F 16/243 (2019.01) [H04L 63/102 (2013.01)] 30 Claims
OG exemplary drawing
 
1. A computer-program product comprising a non-transitory machine-readable storage medium storing computer instructions that, when executed by one or more processors, perform operations comprising:
receiving, via a user interface, a natural language query from a user for executing at least one analytical task;
generating an analytical large language model (LLM) prompt based on the natural language query;
in response to generating the analytical LLM prompt, orchestrating an LLM-directed workflow for handling the natural language query, by:
automatically prompting, using the analytical LLM prompt, an analytical task-oriented LLM to generate at least one structured query for retrieving, from a data catalog application, a recommendation of one or more tables relating to the natural language query and accessible to the data catalog application, wherein the at least one structured query includes a structured request relating to the at least one analytical task associated with the natural language query;
querying the data catalog application using the at least one structured query generated by the analytical task-oriented LLM;
obtaining query results from the data catalog application in response to the at least one structured query, wherein the query results include:
an indication of the one or more recommended tables accessible to the data catalog application, and
metadata associated with the one or more recommended tables;
prompting the analytical task-oriented LLM to identify a given analytical task from a plurality of analytical tasks associated with an analytical service executable by a given analytical agent of a plurality of analytical agents of the analytical service based on an input to the analytical task-oriented LLM of at least a portion of the metadata associated with the one or more recommended tables; and
automatically executing, by the given analytical agent of the plurality of analytical agents, the given analytical task, based on the at least one one or more recommended tables accessible to the data catalog application; and
returning, via the user interface, an output response to the natural language query based on a completion of the given analytical task by the given analytical agent.