Computing systems and methods for controlling agentic artificial intelligence agents

Abstract

A computing system is provided that includes a controller agentic Al agent that communicates with one or more supporting agentic Al agents. The computing system also includes a graphical user interface with a control element that varies a value, and the controller agentic Al agent processes the value to generate one or more controller outputs to control the one or more supporting agentic Al agents.

Description

Atty. Dkt. No. 142937-5001COMPUTING SYSTEMS AND METHODS FOR CONTROLLING AGENTIC ARTIFICIAL INTELLIGENCE AGENTSCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This patent application claims priority to United States Provisional Patent Application No. 63 / 718,537, filed on November 8, 2024, and titled “COMPUTING SYSTEMS AND METHODS FOR RAPIDLY TRAINING AND EXECUTING ARTIFICIAL INTELLIGENCE AGENTS”, and to United States Provisional Patent Application No.63 / 758,296, filed on February 13, 2025, and titled “COMPUTING SYSTEMS AND METHODS FOR CONTROLLING AGENTIC ARTIFICIAL INTELLIGENCE AGENTS”, the entire contents of which are herein incorporated by reference.TECHNICAL FIELD

[0002] The disclosed exemplary embodiments relate to computer-implemented systems and methods for controlling agentic artificial intelligence (Al) agents.BACKGROUND

[0003] In many cases, Al models are beginning to become more common. Larger generative Al models are difficult to fine tune and train for specific applications. In some cases, using Al models can be difficult and may require technical skills to update and fine tune.

[0004] In some cases, training an Al model also uses large amounts of data. In some cases, users do not have access to a sufficient amount of data or the appropriate type of data to develop an effective Al model.

[0005] In some cases, the data is too large and consequently causes various forms of hallucinations.

[0006] In some cases, the user does not know how to write an effective prompt for a large language model (LLM), leading to inaccurate responses.

[0007] In some cases, the user does not how to import, incorporate and generate analysis, outcomes, and recommendations with explainability usingAtty. Dkt. No. 142937-5001 structured data, unstructured data, third party data, and Internet data. In some cases, existing computing systems involve many separate computing software modules and many data sources, which have their own user interfaces, but do not communicate with each other.

[0008] In some cases, a client has their own data that is considered proprietary, confidential or private. Using this client data to train an Al model may create risks for the private data to be released to third parties. In some cases, executing the client data is limited and is not able to be used to train an effective Al model.SUMMARY

[0009] The following summary is intended to introduce the reader to various aspects of the detailed description, but not to define or delimit any invention.

[0010] In some cases, a computing system is provided that includes: a controller agentic Al agent that communicates with one or more supporting agentic Al agents.

[0011] In some cases, the computing system includes a GUI with a control element that varies a value, and the controller agentic Al agent processes the value to generate one or more controller outputs to control the one or more supporting agentic Al agents.

[0012] In some cases, a method includes: providing a GUI comprising a control element operable to control a controller Al agent; responsive to detecting a change of a value of the control element, inputting the value to the controller Al agent; the controller Al agent selecting one or more supporting Al agents based on processing the value; the controller Al agent generating and sending one or more controlling outputs to each of the one or more supporting Al agents; and updating the GUI to display one or more supporting outputs generated by the one or more supporting Al agents.

[0013] In at least a broad aspect, a computing system is provided for controlling artificial intelligence (Al) agents. The computing system comprises: a processor, a communication interface, and memory, the processor coupled to the communication interface and the memory. The computing system further comprise a controller agentic Al agent operable by the processor, and the controller agentic Al agent communicatesAtty. Dkt. No. 142937-5001 with one or more supporting agentic Al agents. The processor is configured to generate a graphical user interface (GUI) comprising a control element that varies a value, and the controller agentic Al agent processes the value to generate one or more controller outputs to control the one or more supporting agentic Al agents.

[0014] In some cases, the processor is further configured to at least: detect a change of the value of the control element and, in response thereof, input the value to the controller agentic Al agent; autonomously select, by the controller agentic Al agent, one or more supporting Al agents based on processing the value; autonomously generate and send, by the controller agentic Al agent, one or more controlling outputs to each of the one or more supporting Al agents; and, update the GUI to display one or more supporting outputs generated by the one or more supporting Al agents.

[0015] In some cases, the controller agentic Al agent receives the one or more supporting outputs from the one or more supporting Al agent; the controller agentic Al agent processes the one or more supporting outputs to generate a controller output; and the processor displays the controller output in the GUI.

[0016] In some cases, the GUI comprises a slider bar element that displays a current value visual element corresponding to a current value, and displays a target value visual element corresponding to a target value; wherein the control element is the target value visual element and the value is the target value; wherein the target value visual element is operably movable along the slider bar element to adjust the target value.

[0017] In some cases, the GUI comprises: a set of elements positioned in relation to each other according to a given relationship; a current value visual indicator identifying a current value associated with the set of elements; and a target value visual indicator that identifies a target value associated with the set of elements; wherein the control element comprises a boundary edge corresponding to the target value visual indicator, and the value is the target value; and, the boundary edge is operably movable to bound a different subset of the set of elements and, in response, vary the target value.

[0018] In some cases, the boundary edge is a digitally visualized lariat operably movable by a user input.Atty. Dkt. No. 142937-5001

[0019] In some cases, the set of elements are positioned in relation to each other according to geographical location.

[0020] In some cases, multiple edges or links are displayed with the set of elements to generate a network graph, wherein the edges or links visually indicate the given relationship.

[0021] In some cases, the processor is further configured to: detect the control element is in in a first state or position in the GUI corresponding to a first instance of the value; subsequently detect an on-focus state of the control element and, responsive to detecting an off-focus state of the control element following the on-focus state, detect that the control element is a second state or position in the GUI; determine a second instance of the value based on the second state or position in the GUI; and processes the second instance of the value to generate the one or more controller outputs to control the one or more supporting agentic Al agents.

[0022] In some cases, the control element comprises a target value visual element that operably moveable along a slider bar element in the GUI.

[0023] In some cases, the control element comprises a digitally visualized lariat operably movable to highlight a subset of elements from amongst a set of elements displayed in the GUI, wherein the set of elements are displayed positioned in relation to each other according to a given relationship.

[0024] In some cases, the controller agentic Al agent or the one or more supporting Al agent, or both, ingest client data from multiple data nodes; wherein the multiple data nodes comprises: a first data node that generates a first output; a second data node that is configured to process at least the first output to generate a second output; and a third data node that is configured to process at least the second output to generate a third output; and, wherein the first output, the second output and the third output form at least part of the client data.

[0025] In some cases, the controller agentic Al agent or the one or more supporting Al agent, or both, ingest client data from multiple data nodes; wherein one or more of the multiple data nodes comprises a node Al agent that is configured to generate node data that forms at least part of the client data.Atty. Dkt. No. 142937-5001

[0026] In some cases, the controller agentic Al agent or the one or more supporting Al agent, or both, ingest client data from multiple data nodes; wherein the multiple data nodes comprise multiple robots.

[0027] In some cases, the controller agentic Al agent or the one or more supporting Al agent, or both, ingest client data from multiple data nodes; wherein the multiple data nodes comprise multiple vehicles.

[0028] In some cases, the controller agentic Al agent or the one or more supporting Al agents, or both, ingest client data from multiple data nodes; wherein the multiple data nodes comprise multiple machines.

[0029] In some cases, the processor is further configured to: ingest, from an external data source, industry-specific data; re-train the fundamental Al model using the industry-specific data to generate an industry-specific Al model; deploy the industry-specific Al model; ingest, from a client data source, client data; re-train the industry-specific Al model using the client data to generate a client Al model; deploy the client Al model for access by the controller agentic Al agent or the one or more supporting Al agent, or both; ingest, from the client data source, new client data; retrain the client Al model using the new client data to generate a new client Al model, and deploy the new client Al model in place of the client Al model for access by the controller agentic Al agent or the one or more supporting Al agent, or both.

[0030] In some cases, the computing system further comprises a client Al computing environment and an Al computing environment separate from the client Al computing environment; wherein the industry-specific Al model is deployed within the client Al computing environment; wherein the client Al model is generated and deployed within the client Al computing environment; wherein the new client Al model is generated and deployed within the client Al computing environment; and wherein the industry-specific Al model is generated in the Al computing environment.

[0031] In at least another broad aspect, a method for controlling artificial intelligence (Al) agents is provided. The method is executed in a computing environment comprising a processor, a communication interface, and memory, the processor coupled to the communication interface and the memory, and the method comprises: establishing a communication between a controller agentic Al agent and one or more supporting agentic Al agents; generating a graphical user interface (GUI)Atty. Dkt. No. 142937-5001 comprising a control element that varies a value; and, processing, using the controller agentic Al agent, the value to generate one or more controller outputs to control the one or more supporting agentic Al agents.

[0032] According to some aspects, the present disclosure provides a non- transitory computer-readable medium storing computer-executable instructions. The computer-executable instructions, when executed, configure a processor to perform any of the methods described herein.BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The drawings included herewith are for illustrating various examples of articles, methods, and systems of the present specification and are not intended to limit the scope of what is taught in any way. In the drawings:FIG. 1 is a schematic block diagram of a computing system with multiple agentic Al agents in accordance with at least some embodiments;FIG. 2 is a schematic block diagram of a computing system showing the flow of data between multiple agentic Al agents in accordance with at least some embodiments;FIG. 3 is a schematic block diagram of an agentic Al agent in accordance with at least some embodiments;FIG. 4 is a graphical user interface (GUI) with a control element to control multiple agentic Al agents in accordance with at least some embodiments;FIG. 5 is a GUI in a subsequent instance of FIG. 4 in accordance with at least some embodiments;FIG. 6 is a GUI in a subsequent instance of FIG. 4 in accordance with at least some embodiments;FIG. 7 is a flowchart diagram of an example method of controlling one or more agentic Al agents, in accordance with at least some embodiments;FIG. 8 is a schematic block diagram of another computing system with multiple agentic Al agents in accordance with at least some embodiments;Atty. Dkt. No. 142937-5001FIG. 9 is a schematic block diagram of another computing system with multiple agentic Al agents in accordance with at least some embodiments;FIG. 10 is a schematic block diagram of another computing system with multiple agentic Al agents in accordance with at least some embodiments;FIG. 11 is a schematic block diagram of another computing system with multiple agentic Al agents in accordance with at least some embodiments;FIG. 12 is a schematic block diagram of a computer, in accordance with at least some embodiments;FIG. 13A is another GUI with a control element in a first state to control multiple agentic Al agents in accordance with at least some embodiments; andFIG. 13B is the GUI shown in FIG. 13A with the control element in a second state to control multiple agentic Al agents in accordance with at least some embodiments.DETAILED DESCRIPTION

[0034] In some cases, a computing system is provided that includes multiple agentic Al agents that are configured to ingest data. In some cases, each agentic Al agent includes or uses large language models (LLMs) or small language models (SLMs), or both, and analytic modules to present natural language outcomes based on ingested data, recommendations with explainability, and next steps. In some cases, the ingested data includes structured data and unstructured data from third-party data sources. In some cases, the analytic modules include a surface module to identify relevant data from ingested data, a trend module to determine patterns and trends in the ingested data, a recommendations module to generate recommendations, an infer module to generate inferences from the ingested data, or a predict module to compute predictions based on the data from the other modules (e.g., relevant data, trend data, recommendation data, inferred data, internal data sources, external data sources), or a combination of these modules. In some cases, the analytic modules include a heuristics modules, which uses heuristic computations to determine data features. In some cases, the analytics modules include a network graph module that visualizes actual and predicted edge I node relationships in static modes, hybrid human Al agent interactive dynamic modes, and time series modes.Atty. Dkt. No. 142937-5001

[0035] In some cases, the trend module is configured to computed machine learning computations to determine trends and forecasts.

[0036] In some cases, the recommendations module is configured to execute Strengths Weakness Opportunities and Threats (SWOT) analysis and provide one or more recommendations based on the SWOT analysis. For example, the recommendations module includes a LLM or a SLM. The computing system automatically generates a prompt that specifies conducting the SWOT analysis using internal data and / or external data, and providing one or more recommendations based on the SWOT analysis. The computing system then automatically inputs the prompt into the LLM or SLM of the recommendations module.

[0037] In some cases, the inference module is configured to obtain features of comparative entities, and compute look-a-like computations to identify target entities that are similar to the comparative entities. In some cases, the look-a-like computations include using cosine similarity computations.

[0038] In some cases, the predict module predicts supply and demand, also called supply and demand forecasting.

[0039] In some cases, each agentic Al agent processes and displays outcomes and recommendations with explainability. In some cases, each agentic Al agent operates operate independently on its own without external prompts. In some cases, each agentic Al agent is autonomous and operates independently. In some cases, each agent Al agent includes one or more goals and executes reasoning, analytics and actions to complete the one or more goals. In some cases, each agent Al agent has persistent memory and continuous learning capabilities to adjust its own functions, parameters, hyperparameters, and data.

[0040] In some cases, an agentic Al agent is configured to select different Al agent(s) to adopt and process data differently based upon different signals, different Al agent outcomes, and or threshold changes in the exogenous data requiring a different Al to process the data.

[0041] In some cases, each agentic Al agent is configured to process complex operations in computing environments with a high degree of variability. In some cases, each agentic Al agent generates its own decisions based on its own reasoning and analysis. In some cases, each agentic Al agent is configured to interact with and adaptAtty. Dkt. No. 142937-5001 to external computing resources, including external databases, external platforms, external computing devices (e.g., client devices, third party servers, etc.), and external agentic Al agents. In some cases, each agentic Al agent is configured to modify one or more of its own primary goals or one or more of its own sub-goals. In some cases, a sub-goal is data linked to a primary goal and helps to accomplish the primary goal.

[0042] In some cases, an agentic Al agent automatically searches for other agentic Al agents and connects with those other agentic Al agents. In some cases, a data connection (or data connections) between a given agentic Al agent and the one or more other agentic Al agents includes sending requests for updates from the one or more other agentic Al agents, and automatically providing data to those upstream / downstream agentic Al agents. In some cases, in turn, each of given Al agentic agent and the one or more other agentic Al agents generate outputs that include recommendations and explainability in natural language.

[0043] In some cases, a hybrid a human and a Al agent interact to provide more domain and range specificity for problem solving. The human may ask specific natural language from the Al agent recommendations. Based on these human queries, the human may fine tune the algos, machine learning, prompts, modules, foundation models, or combinations of the aforementioned. The Al agent may ask specific natural language questions to the human seeking guidance based on certain thresholds, event situations, business rules, situations, conditions, or combinations of the aforementioned.

[0044] In some cases, the computing system includes multiple agentic Al agents that are serially data linked and sequentially process data within an Al framework of the computing system. In some cases, each agent Al agent in the serial configuration generates and transmits outputs that include recommendations with explainability in natural language. In some cases, explainability includes uniform resource locator (URL) links to internal and external data sources providing further support to the recommendation. In some cases, explainability includes actual transactions, metrics, and key performance indicators (KPIs) providing further support to the recommendation. In some cases, explainability includes links to internal or external source of truth documents providing further support to the recommendation.Atty. Dkt. No. 142937-5001

[0045] In some cases, one agentic Al agent simultaneously transmits data to multiple serially data linked agentic Al agents. In some cases, the transmission of data from the one agentic Al agent to the multiple serially data linked agentic Al agents occurs simultaneously (or near simultaneously), so that the multiple serially data linked agentic Al agents receive the data at the same time.

[0046] In some cases, an agentic Al agent automatically searches and work with external Al agents, including external agentic Al agents. For example, the agentic Al agent of the computing system is able to interact with an external agentic Al agent operating as part of a third-party platform. For example, querying and capturing weather data sources, traffic report conditions, sick and flu / virus outbreaks, regional / local events all of which are published and available 3rd party Internet data. These data sources and the consequential impacts can subsequently be geo mapped and overlaid with the internal Al agents to provide a multi-data view combining actual, predicted, and pop up exogenous data. In some cases, this interaction includes querying specific data from an external data source that is specific and germane to the content that a given agentic Al agent (of the computing system) needs to deliver as a standalone agent outcome and recommendation with explainability, and next steps. In some cases, this data provides both primary and secondary data to support a specific agentic Al agent standalone nature of delivering relevant outcomes to a client data account or to a downstream I upstream agentic Al agent.

[0047] In some cases, there is a hybrid, human and Al agent interaction computing system GUI that narrowly defines the domain and range of knowledge, data, querying and generating specific responses and or outcomes based upon this hybrid interaction.

[0048] In some cases, there is a hybrid, human and Al agent interaction computing system GUI that enables the user to use different methods, including but not limited to type ahead, autofill, and or explicitly input pre-defined user defined prompts facilitating faster and easier interaction between the human and the Al agents.

[0049] In some cases, there is a hybrid, human and Al agent interaction computing system GUI that enables the user to capture and incorporate data (structured and unstructured) data residing outside of an internal system and / or external 3rd party / Internet data sourcesAtty. Dkt. No. 142937-5001

[0050] In some cases, there is a hybrid, human and Al agent interaction computing system GUI that enables the user to input a human readable user defined function object from a library of predefined objects. Examples of these user defined objects include user defined prompts, user defined visualizations, user defined algorithms, user defined internal and external data sets, and user defined 3rd party data sources.

[0051] Referring to FIG. 1 , a computing system 100 is provided that includes multiple agentic Al agents. An Al computing environment 110 communicates with one or more external data sources 150 to ingest various types of data.

[0052] The Al computing environment 110 includes a datastore 112 that stores one or more controller Al models 114 and one or more supporting Al models 116. In some cases, these models include large language model or small language models, or both. In some cases, these models include other types of Al models, including generative Al models, generative adversarial networks (GANs), recurrent neural networks (RNNs), retrieval augmented generation models (RAGs), other artificial neural networks, machine learning models, and Surface-Trend-Recommend-Infer- Predict-Action (STRIPA) models.

[0053] In some cases, the Al computing environment 110 includes an agentic Al compute environment 118 that includes a controller agentic Al agent 120, and one or more supporting agentic Al agents. In some cases, there is a supporting agent Al agent for Domain A 121 , a supporting agent Al agent for Domain B 122, and a supporting agent Al agent for Domain C 123. Domain A, Domain B and Domain C are separate areas of knowledge, or activity, or operation, or accounts, or other data division. In some cases, the controller agentic Al agent 120 transmits data to one or more of the supporting agentic Al agents, and the one or more supporting agentic Al agents execute compute operations and analytics. In some cases, outputs generated by the supporting agentic Al agents are sent back to the controller agentic Al agent. In some cases, outputs generated the supporting agentic Al agents are not sent back to the controller agentic Al agent, and used by other systems, apps, or functions. In some cases, the supporting agentic Al agent, responsive to the data provided by the controller agentic Al agent, executes actions. In some cases the supporting agentic Al agents transmits the results, outcomes, recommendations with explainability with 3rd party entities or ecosystems providing real time / near real time updates.Atty. Dkt. No. 142937-5001

[0054] In some cases, the controller agentic Al agent 120 is loaded with, and includes, a given controller Al model 114. In some cases, the controller agentic Al agent 120 communicates with a given controller Al model 114. In some cases, the controller agentic Al agent 120 is able to swap or switch between different controller Al models (e.g., replace a currently loaded or currently connected controller Al model with a different controller Al model).

[0055] In some cases, the controller agentic Al agent 120 intelligently swaps / switches based upon different modes. For example, a supporting agent detects a change in data, trends, values, exogenous data and automatically determines the existing model (e.g., STRIPA model, ML / RNN model, forecast model, etc.) is no longer effectively modeling / simulating / predicting within a certain operating condition, and the supporting agent automatically runs simulations to determine if a different model is better able to predict / operate given the change in operating conditions. The controller agentic Al agent 120, in response to identifying a higher performing Al model, automatically switches / swaps to that higher performing Al model.

[0056] In another example for intelligently swapping / switching, the controller agentic Al agent prompts a human, in a hybrid human / AI agent mode, providing a STAR (Situation, Task, Action, Result) situation to the human, stating the operating condition, and providing recommendations with explainability to the user. Thereafter, the user provides guidance I directives to swap / change to a different model. In response, the controller agentic Al agent uses the guidance / directives obtained from the user interact to accordingly replace a currently loaded or currently connected controller Al model with a different controller Al model.

[0057] In some cases, the controller agentic Al agent 120 is able to modify a controller Al models, including a currently loaded or currently connected controller Al model, or a currently unloaded or currently unconnected controller Al model.

[0058] In some cases, a supporting agent Al agent is loaded with, and includes, a given supporting Al model 116. In some cases, the supporting agentic Al agent communicates with a given supporting Al model 116. In some cases, the supporting agentic Al agent is able to swap or switch between different supporting Al models (e.g., replace a currently loaded or currently connected supporting Al model with a different supporting Al model). In some cases, the supporting agentic Al agent is ableAtty. Dkt. No. 142937-5001 to modify a supporting Al model, including a currently loaded or currently connected supporting Al model, or a currently unloaded or currently unconnected supporting Al model.

[0059] In some cases, the controller agentic Al agent uses the outputs from one or more supporting Al agents to generate further outputs. In some cases, these further outputs are transmitted to one or more supporting agentic Al agents to execute further computations. In some cases, these further outputs are transmitted to other systems, apps, or functions.

[0060] In some cases, one or more of the agentic Al agents communicate with other compute resources, including one or more Al agents 132, one or more function modules 134, and one or more database 136. In some cases, these other compute resources are part of a supporting data layer 130. In some cases, the agentic Al agents communicate with internal and external knowledge and data foundation layers. In some cases, the agentic Al agents communicate with internal and external knowledge and data foundation layers.

[0061] In some cases, the agentic Al computing environment 118 communicates with an app 126 (or application) that includes a graphical user interface (GUI) 128. In some cases, the GUI 128 includes a control element that is used to interact with the controller agent Al agent 120.

[0062] In some cases, a client system 140 interacts with the agentic Al compute environment 118. For example, one or more the agentic Al agents in the agentic Al computing environment 118 interact with the client system 140. In some cases, the client system interacts with the Al computing environment 110. In some cases, the client system 140 includes one or more client data sources 142 and one or more client user computing devices 144. In some cases, these user computing devices include on-prem devices and off-prem devices. In some cases, the user computing devices and or Al agents are autonomously operating in an ecosystem consisting of upstream internal and external devices, Al agents, people, and ecosystems, within the host Al platform ecosystem of agentic Al agents and knowledge / data layers, and with downstream internal and external, Al agents, people, and ecosystems. Examples of user computing devices include desktop computers, tablets, laptops, smartphones,Atty. Dkt. No. 142937-5001 computing kiosks, computers within a vehicle, wearable computers (e.g., computing headsets, goggles, smart headphones, and smart watches), etc.

[0063] In some cases, one or more external agentic Al agents 162 communicate with one or more agentic Al agents located in the Al computing environment 110. The external Al agent 162 is part of an external system 160. In some cases, the external Al agent 162 part for a platform that includes one or more of the external data sources 150.

[0064] Referring to FIG. 2, a system diagram 200 shows communications between various components.

[0065] In some cases, the GUI 128 includes one or more control elements 202 that generate outputs that are transmitted to the controller agentic Al agent 120. The controller agentic Al agent processes the output from the control element 202 to generate one or more outputs that are transmitted to one or more supporting agentic Al agents. In some cases, the controller agentic Al agent processes the output from the control element 202 to determine its own outputs, and to determine which one or more supporting agent Al agents are to be affected by its own outputs. In an example case, in first instance, the controller agentic Al agent processes an output from the control element 202 to determine its own output that is only sent to supporting agentic Al agent for Domain A 121 ; and in second instance, the controller agentic Al agent processes a different output from the control element 202 to determine its own different outputs that are sent to the supporting agentic Al agent for Domain A 121 , the supporting agentic Al agent for Domain B 122, and the supporting agentic Al agent for Domain C 123.

[0066] In some cases, supporting agentic Al agents use the output from the controller agentic Al to generate their own results and actions. These results could include computing analytics, generating recommendations, generating predictions, generating inferences, or executing actions, or a combination thereof. In some cases, the supporting agentic Al agents communicate with one or more other Al agents, function modules, databases, external agentic Al agents, external data sources, or the client system, or a combination thereof.

[0067] In some cases, a supporting agentic Al agent observes and / or ingests data from one or more other Al agents, function modules, databases, external agenticAtty. Dkt. No. 142937-5001Al agents, external data sources, or the client system, or a combination thereof. The supporting agentic Al agent then executes computations that generate a result that is pushed to the controller agentic Al agent 120. In other words, a supporting agentic Al agent acts autonomously and can transmit data, including prompts, recommendations, predictions, trends, and actions to the controller agentic Al agent 120, without the controller agentic Al agent 120 sending any prior prompts or data that would trigger a response from a supporting agentic Al agent.

[0068] In some cases, a supporting agentic Al agent for Domain A 121 communicates with one or more sub-supporting agentic Al agents (e.g., subsupporting agentic Al agent for Domain A.1 204 and sub-supporting agentic Al agent for Domain A.2 206). In some cases, Domain A.1 and Domain A.2 are sub-domains of Domain A. Data is exchanged between the sub-supporting agentic Al agents and the supporting agentic Al agents.

[0069] In some cases, the controller agentic Al agent receives results, actions, recommendations, trends, predictions, data, etc. from one or more of the supporting agentic Al agents, and then generates its own controller output that includes results, actions, recommendations, trends, predictions, or data, or combination thereof. The controller output, in some cases, is sent to the GU1 128 for display. In some cases, the controller output is used to automatically initiate an action in other components, or external computing system, or the client system. In some cases, the action occurs or is executed by upstream and downstream humans, devices, Al agents, computing devices and applications, or vendor / client systems, or a combination thereof. Furthermore, these aforementioned may communicate back to the originating controller agentic Al agent in a two-way messaging process to provide explainability and traceability (e.g., via a report or a system audit log) of autonomous, human, or hybrid human agentic Al actions.

[0070] In some cases, a supporting agentic Al agent communicates with the GUI 128, so that the GUI displays outputs from the supporting agentic Al agent.

[0071] Referring to FIG. 3, an agentic Al agent 300 is provided. In some cases, this structure for an agentic Al agent 300 is applied to the controller agentic Al agent 120, the supporting agent Al agent for Domain A 121 , the supporting agent Al agent for Domain B 122, and the supporting agent Al agent for Domain C 123.Atty. Dkt. No. 142937-5001

[0072] The agentic Al agent 300 includes an agentic Al agent engine 302 that communicates with a search and ingest module 330, memory 320, one or more LLMs 340 and / or small language models, and one or more functional modules 350.

[0073] The search and ingest module 330 searches for and / or ingests data from other data sources and Al agents.

[0074] The memory 320 stores thereon goals, sub-goals, reasoning data, planning data, coordinating data, and action data. This data persists in the memory.

[0075] In some cases, the agentic Al agent engine 302, or “engine”, includes a perception module for pre-processing data; a reasoning module for processing the data to generate analytics and / or to generate explanations; a planning module for generating plans that include coordination and / or actions; a coordination module to coordinate actions; and / or an action module to execute the actions. In some cases, these modules obtain data from the memory 320 to execute the operations.

[0076] In some cases, the reasoning module includes a surface module, a trend module, a recommend module, an infer module, and a predict module.

[0077] In some cases, the predict module includes a forecast module configured to generate forecasts and a promotion module configured to suggest autonomous product, price, place, and promotion (4Ps) recommendations. In some cases, the promotion module autonomously analyses firmographic, demographic, and exogenous data, and corresponding forecasted data from the forecast module, to generate the 4Ps recommendations. In some cases, the forecast module computes forecasts for stock availability and the promotion module uses the forecasts to generate a new old stock (NOS) autonomous recommendation to sell inventory that is new but old and apply the 4Ps to sell and get the inventory off the books. In some cases, the forecast module includes a look-a-like I cosine similarity sub-module to model and forecast new products using the 4Ps recommendations to help model and forecast new product introductions based on prior customer, vendor, firmographic and demographic similar historical structured and unstructured data and machine learnings.

[0078] In some cases, the modules of the agentic Al agent engine use the one or more LLMs 340, or one or more functional modules 350, or both, to execute operations.Atty. Dkt. No. 142937-5001

[0079] In some cases, the control element 202 in the GUI is a slider bar. In some cases, the control element 202 includes one or more buttons. In some cases, the control element 202 is a text interface. In some cases, the control element 202 includes an audio interface that receives voice data or other audio data. In some cases, the control element 202 includes an image recognition interface that processes imagery.

[0080] In some cases, the control element can take the form of a network graph and a "grabber" or digitally visualized lariat to circumscribe an area to simulate and make network optimization recommendations with explainability. In some cases, control element can take the form of a demand forecast and a "grabber1or digitally visualized lariat to circumscribe an area to simulate a forecast in a certain domain and range for existing or new product introductions with explainability. In some cases, the control element can take the form of a grabber or digitally visualized lariat and circumscribe the domain and range of historical data to dig deeper into what historically happened with explainability. In some cases, the control element can take the form of a grabber or digitally visualized lariat and circumscribe the domain and range of overlapping historical and predicted data to dig deeper into how the actual versus predicted model performed and provide recommendations to further fine tine the prediction module agentic Al agents.

[0081] Referring to FIG. 4, an example GUI 400 is shown that includes a slider bar 402, which is a type of control element 202.

[0082] The slider bar 402 shows a current value, including a current value visual element 406, and a target value, including a target value visual element 408, which spans between the current value and the target value. The difference between the current value and the target value is W. A slide control element 404 slides along the slider bar 402, to adjust the difference between the current value and the target value.

[0083] An information area 409 shows results generated by a controller agentic Al agent, a supporting agentic Al agent for Domain A, a supporting agentic Al agent for Domain B, and a supporting agentic Al agent for Domain C.

[0084] The controller agentic Al agent 120 has a goal to achieve an increase in value of W, from the current value to the target value, based on the data provided from the slider bar 402. The controller agentic Al agent processes data to achieve this goal,Atty. Dkt. No. 142937-5001 include interacting with one or more of the supporting agentic Al agents from different domains.

[0085] The supporting agentic Al agent for Domain A 121 generates and outputs a summary statement for Domain A 410, and detailed action for Domain A 411 , which achieve an increase in value of X. In some cases, this recommendation and calculation is coordinated and adjusted by the controller agentic Al agent 120, so that the cumulative increase in value from amongst the recommendations and / or actions of the different domains is W.

[0086] The supporting agentic Al agent for Domain B 122 generates and outputs a summary statement for Domain B 414, and detailed action for Domain B 415, which achieve an increase in value of Y. In some cases, this recommendation and calculation is coordinated and adjusted by the controller agentic Al agent 120, so that the cumulative increase in value from amongst the recommendations and / or actions of the different domains is W.

[0087] The supporting agentic Al agent for Domain C 123 generates and outputs a summary statement for Domain C 418, and detailed action for Domain C 419, which achieve an increase in value of Z. In some cases, this recommendation and calculation is coordinated and adjusted by the controller agentic Al agent 120, so that the cumulative increase in value from amongst the recommendations and / or actions of the different domains is W.

[0088] The values of X + Y +Z is equal to W. As the value of W changes based on the position of the slide control element 404, the controller agentic Al agent automatically updates itself and coordinates with the one or more supporting agentic Al agents to generate updated analysis, recommendations and actions to achieve the new value of W.

[0089] In some cases, the GUI 400 also includes an analysis 420 for Domain A.1 , a sub-domain of Domain A. The analysis 420, in some cases, is provided by the sub-supporting agentic Al agent for Domain A.1 204.

[0090] In some cases, the GUI 400 also includes an analysis 422 for Domain A.2, a sub-domain of Domain A. The analysis 422, in some cases, is provided by the sub-supporting agentic Al agent for Domain A.2 206.Atty. Dkt. No. 142937-5001

[0091] The analysis 420 and the analysis 422 provide additional supporting explanation for the recommendation and / or detailed action for Domain A 411 .

[0092] FIG. 5 shows another GUI 500, which is a subsequent instance of the GUI 400 after the slide control element 404 has been moved to a decreased position, showing a projected value. The projected value is between the target value and the current value. The difference between the projected value and current value is V, which is less than W.

[0093] The controller agentic Al agent and the supporting agentic Al agents automatically recompute to generate recommendations and / or actions in the different domains. The summary statement and actions for each domain are updated, as well as the contributing value for the increase. The sum of the contributing value S in Domain A, the contributing value T in Domain B, and the contributing value U in Domain C is V.

[0094] In some cases, the computing system detects the control element (e.g., the slide control element 404) is in in a first state or position in the GUI corresponding to a first instance of a value (e.g., the target value). The computing system subsequently detects an on-focus state of the control element, for example, during which the control element is moved by user input. Responsive to detecting an off-focus state of the control element following the on-focus state, the computing system detects that the control element is in a second state or position in the GUI. For example, the slide control element 404 has been moved along the slider bar 402. The computing system determines a second instance of the value (e.g., the target value) based on the second state or position in the GUI, and processes the second instance of the value to generate the one or more controller outputs to control the one or more supporting agentic Al agents. In some cases, the second instance of the value, or the change of the value, is input into the controller agentic Al agent. The controller agent Al agent then autonomously selects one or more supporting Al agents based on processing the value. The controller agentic Al agent then autonomously generates and sends one or more controlling outputs to each of the one or more supporting Al agents. The one or more supporting Al agents then process the one or more controlling outputs to generate one or more supporting outputs. Responsive to obtaining the one or more supporting outputs from the one or more supporting Al agent, the computing system updates the GUI to display the one or more supporting outputs.Atty. Dkt. No. 142937-5001

[0095] FIG. 6 shows another GUI 600, which is a subsequent instance of the GUI 400 after the slide control element 404 has been moved to a decreased position, showing a projected value. The projected value is between the target value and the current value. The difference between the projected value and current value is V, which is less than W. In this example, the controller agentic Al agent 120 determines to only use recommendations and / or actions to from supporting agentic Al agent for Domain A 121 to achieve the increase in value V. The supporting agentic Al agent for Domain A 121 may generate different recommendations and / or actions that would accomplish the increase in value V, compared to the summary 210 and the action 411 in the instance of FIG. 4.

[0096] Turning to FIG. 13A, another GUI 1300 is shown that includes some similar features to FIG. 4. Contextual data is shown by displaying various visual elements 1304 in spatial relation to each other. The elements 1304 could represent data nodes, vendors, devices, entities, cities, etc. In some cases, the elements 1304 are positioned according to geography or location (e.g., according to a map). In some cases, the elements 1304 are positioned relative to each other according to logistical relationship, or hierarchical relationship, or communication relationship. In some cases, edges or links are displayed between certain elements 1304 to generate a network graph. In some cases, the elements 1304 are displayed using the same visual icon. In some other cases, the elements 1304 are displayed using different visual icons.

[0097] The GUI 1300 also computes and displays a current value visual indicator 1306 identifying a current value associated with the set of elements 1304. The current value visual indicator 1306, for example, highlights or bounds a subset of the set of elements, indicating a current set of elements are identified according a given threshold or parameter.

[0098] The GUI 1300 further displays a target value visual indicator 1308 that identifies a target value associated with the set of elements 1304. In some cases, the target value visual indicator 1308 is different from the current value visual indicator 1306 and highlights or bounds a secondary subset of the set of elements, associated with a target value. For example, the secondary subset of the set of elements could include more elements, or less elements, or a different set of elements compared to the subset of the set of elements highlighted by the current value visual indicator 1306.Atty. Dkt. No. 142937-5001

[0099] In the example in FIG. 13A, the target value visual indicator 1308 highlights or bounds the same subset of the set of elements as the current value, and includes additional elements.

[0100] The entire set of elements 1304 shows context data.

[0101] The GUI 1300 provides a control element, which is also the target value visual indicator 1308, that can be moved via a grabber input to change the boundary. This is a form of a digitally visualized lariat. Other types of visual control elements can be used.

[0102] The GUI 1300 also includes the information area 409, and the analyses 420 and 422, which vary depending on the target value and the current value. The embodiment shown in FIG. 13A is representative of a first state of a current value and a given target value, for example at a time t1 . A user redraws (e.g., click and drag) the control element (e.g., the edge of the target value visual indicator 1308) to bound or highlight a different target subset of elements (e.g., forming a different target value).

[0103] Turning to FIG. 13B, the GUI 1300 is shown in a second state (e.g., at a subsequent time t2) of the current value and a different given target value. The data in the information area 409, and the analyses 420 and 422, are subsequently updated as a result of the different given target value.

[0104] In some cases, the computing system detects the control element (e.g., the target value visual indicator 1308) is in in a first state or position in the GUI corresponding to a first instance of a value (e.g., the target value). The computing system subsequently detects an on-focus state of the control element, for example, during which the control element is moved by user input. Responsive to detecting an off-focus state of the control element following the on-focus state, the computing system detects that the control element is in a second state or position in the GUI. For example, the target value visual indicator 1308 has been moved to redraw a boundary around a different subset of elements from amongst the set of elements 1304. The computing system determines a second instance of the value (e.g., the target value) based on the second state or position in the GUI, and processes the second instance of the value to generate the one or more controller outputs to control the one or more supporting agentic Al agents. In some cases, the second instance of the value, or the change of the value, is input into the controller agentic Al agent. The controller agentAtty. Dkt. No. 142937-5001Al agent then autonomously selects one or more supporting Al agents based on processing the value. The controller agentic Al agent then autonomously generates and sends one or more controlling outputs to each of the one or more supporting Al agents. The one or more supporting Al agents then process the one or more controlling outputs to generate one or more supporting outputs. Responsive to obtaining the one or more supporting outputs from the one or more supporting Al agent, the computing system updates the GUI to display the one or more supporting outputs.

[0105] Turning to FIG. 7, an example method 700 is provided.

[0106] Block 702: Provide a Controller Al Agent that is linked to multiple Supporting Al Agents.

[0107] Block 704: Provide a GUI that comprising a control element operable to control the Controller Al Agent.

[0108] Block 706: Responsive to detecting an input of the control element, inputting the input to the Controller Al Agent.

[0109] Block 708: Controller Al Agent selects one or more Supporting Al Agents based on processing the input.

[0110] Block 710: Controller Al Agent generates and sends specific supporting inputs for each selected Supporting Al Agent.

[0111] Block 712: Each selected Supporting Al Agent processes the respective supporting input to generate a respective supporting output.

[0112] Block 714: Controller Al Agent receives supporting outputs from the selected Supporting Al Agents and uses same to generate a Controller output.

[0113] Block 716: Display the Controller output in the GUI.

[0114] Block 718: Display supporting outputs from each selected Supporting Al Agent in the GUI.

[0115] In some cases, block 718 occurs independently from block 714 and 716.

[0116] In some cases, a computing system is provided that includes: a controller agentic Al agent that communicates with one or more supporting agentic Al agents. In some cases, the computing system includes a GUI with a control elementAtty. Dkt. No. 142937-5001 that varies a value, and the controller agentic Al agent processes the value to generate one or more controller outputs to control the one or more supporting agentic Al agents.

[0117] In some cases, a method includes: providing a GUI comprising a control element operable to control a controller Al agent; responsive to detecting a change of a value of the control element, inputting the value to the controller Al agent; the controller Al agent selecting one or more supporting Al agents based on processing the value; the controller Al agent generating and sending one or more controlling outputs to each of the one or more supporting Al agents; and updating the GUI to display one or more supporting outputs generated by the one or more supporting Al agents.

[0118] In some cases, the control element is a slide bar. In some cases, the control element is a digital grabber. In some cases, the control element is digitally visualized lariat. Other types of visual control elements can be used to control the controller Al agent.

[0119] Referring to FIG. 8, in some cases, data from the external sources is used to train a fundamental Al model 812. In some cases, industry specific data is used to train one or more industry application Al models. For example, a trained industry application A (or industry specific) Al model 814 is relevant to a particular industry A, and a trained industry application B (or industry specific) Al model 816 is relevant to a particular industry B. For example the industries A and B are different industries. Some examples of different industries include: supply chain industry, manufacturing industry, oil and gas industry, logging industry, data security industry, transportation, logistics, warehousing, NAICS vendors and customers, industry / sub industry ecosystems, etc. In some cases, an Al model library stores multiple trained Al models that trained to be specific to different industries and / or industry applications.

[0120] In some cases, data ingested from the external data sources is used to build and develop knowledge and data foundation layers that are applicable to a given industry. The ingested data from the external sources is converted by an Al model (which may be the fundamental Al model or an indexing Al model) to vector data. The vector data is stored in a vector database in the Al computing environment 110, as better shown in FIGs. 10 and 11 . Storing vector data in a vector database reduces the amount of data that is stored. In some cases, the storing vector data in vector databaseAtty. Dkt. No. 142937-5001 reduces the number of computation steps for the Al model when retrieving the vector data and processing the same by a fundamental Al model and / or an industry application Al model.

[0121] In the example in FIG. 8, a client Al computing environment 818 is provided that is dedicated to a given (e.g., Client 1 ). The client Al computing environment 818 communicates with client 1 systems (e.g., client 1 data sources and client 1 user computing devices).

[0122] In some cases, the client 1 Al computing environment 818 includes multiple agenticAI agents, having the same or similar architecture as the Al computing environment 110.

[0123] In operation, the fundamental Al model is trained using data and analytics from many different data sources, to build a general capability and understanding. The data and analytics include structured data, unstructured data, video, pictures, sound, reports, contracts, policies, natural language text, or industry specific generated KPIs and metrics, or a combination thereof.

[0124] One or more industry application Al models are trained to conduct analysis and execute logical conditions based on industry specific data. This creates a more specialized Al model for a given industry. A client within the same given industry retrains the industry application A trained Al model using their own data sources (e.g., Client 1 data sources) to more quickly train and refine a client specific Al model (e.g., Client 1 Al model). In this way, Client 1 ’s Al model is quickly trained. From the perspective of Client 1 , only Client 1’s data is used to train the Client 1 Al model.

[0125] In some cases, the client Al model is trained using control data that comprises natural language data, such as found in word documents, PDFs, videos, audio data, images, presentation slides, etc. In some cases, the control data includes contracts, policies, methodologies of thinking, guidelines, standard operating procedures, rules, statements of work, regulations, etc. During the training process, the control data is indexed and stored as vectors in a vector database. In some cases, the client Al model extracts natural language logic conditions from the control data and stores the same, or embeds the same within the Al model. In some cases, the naturalAtty. Dkt. No. 142937-5001 language logic conditions are natural language prompts specific to the industry of the client and specific to the control data provided.

[0126] After the trained client Al models are deployed, they are executed using live data that is provided, in some cases, by client data sources. In some cases, the client data sources include multiple data nodes. In some cases, the client data sources include knowledge and data stores, and foundation layers.

[0127] For example, in a supply chain industry, there a first data node represents a manufacturer of a product; a second data node represents a packing compony for packing the product; a third data node represents a shipping company for shipping the packaged product; a fourth data node represents a warehousing and distribution company for temporarily storing the packaged product before distributing the packaged product to stores; and multiple instances of a fifth data node represents multiple stores that receive the packaged product and sell the packaged product. These data nodes are within a process ecosystem of a given client. Data from each of the data nodes can provided to the client Al model to determine scoring according to the natural language logic conditions. For example, the client Al model can determine whether the different vendors of the different data nodes is compliant or non-compliant with any contracts, policies, methodologies of thinking, guidelines, standard operating procedures, rules, statements of work, etc. which form the control data.

[0128] In another example, a first data node represents a manufacturer of a product; upstream from this manufacturer there can be N number of data nodes such as raw material suppliers and subcontractor manufacturers that make components, which are in turn used by the final manufacture at the aforementioned first data node. A second data node represents N number of nationwide wholesalers or third-party logistics (3PL) warehouse firms that stock inventory for subsequent channel distribution to (a) smaller regional / local distributors; a third data node is the actual store(s) that sell the product to the businesses and customers. Between each of these nodes are N number of transportation data nodes representing the shippers moving product from raw materials and subcontractor component manufacturers to the primary manufacturer to the final manufacturer to downstream wholesalers, 3PLs, and ultimately the buyers. The ultimate buyers could have further downstream warehouse, logistics, and transportation data node. These aforementioned data nodes simplistically describe and define the technology ecosystems for a given businessAtty. Dkt. No. 142937-5001 involved in a supply chain. Depending on where a business sits in the aforementioned ecosystem affects the importance of certain data nodes. Data from each of the data nodes can provided to the client Al model to determine scoring according to the natural language logic conditions. For example, a client Al model can determine whether the different vendors of the different data nodes is compliant or non-compliant with any contracts, policies, methodologies of thinking, customer and vendor contracts, internal business guidelines, government I legal compliance, standard operating procedures, business rules, statements of work, etc. which form the control data.

[0129] This also applies to other industries, such as manufacturing. For example, the data nodes represent different manufacturing stations within a manufacturing process, and each manufacturing station produces data.

[0130] Turning to FIG. 9, the system 900 shows multiple data nodes 904a, 904b, 904c that are part of the Client 1 data sources. These data nodes are, in some cases, part of Client 1 ’s ecosystem process. In some cases, each data node transmits data to the client 1 Al computing environment 818, which is processed by the agentic Al agents.

[0131] In some cases, the data provided by the data nodes includes one or a combination of: documents, video data, image data, audio data (e.g., speech data, or environmental audio data, or music, etc.), natural language data, structured data, knowledge and industry specific foundation layers, machine data, sensor data, and platform data (e.g., for Enterprise Resource Planning (ERP) platform, analytics platform, etc.). In some cases, different data nodes provide different types of data.

[0132] In some cases, each of the data nodes 904a, 904b, 904c respectively include their own node Al agent(s). The node Al agents locally process data on the data node, using edge computing. In some cases, the node Al agents include and run small language models (SLMs). In some cases, the node Al agents communicate with the controller agentic Al agent 120, or a supporting agentic Al agent (e.g., 121 , 122, 123), or an Al agent 132, or a function module 134, or a database 136, or a combination thereof.

[0133] In some cases, the data nodes are robotic devices configured to communicate with the Al computing environment 110 and, in some cases, also store and run a node Al agent. In some cases, the robotic devices are humanoid robots.Atty. Dkt. No. 142937-5001

[0134] In some cases, the data nodes are machines configured to communicate with the Al computing environment 110 and, in some cases, also store and run a node Al agent. In some cases, the machines are computer numerically controlled machines. In some cases, the machines are 3D printers. In some cases, the machines include a series or network of material handler machines (e.g., pick and grab machines, conveyor machines, automated guided vehicles, autonomous mobile robots, stacker machines, etc.).

[0135] In some cases, the data nodes are vehicles configured to communicate with the Al computing environment 110 and, in some cases, also store and run a node Al agent. In some cases, the vehicles are autonomous vehicles. In some cases, the vehicles are unmanned vehicles. In some cases, the vehicles are a fleet of transport vehicles. In some cases, the vehicles are aerial drones. In some cases, the vehicles are marine drones. In some cases, the vehicles are shipping vessels. In some cases, the vehicles are trains. In some cases, the vehicles are trucks, such as transport trucks. In some cases, the vehicles are space vehicles, such as satellites.

[0136] In some cases, the Client 1 Al model is executed by the Client 1 Al Agent to process the data from the data nodes, and to output a response to a query, to output an alert, or to output a report, or to output platform data, or a combination thereof.

[0137] FIG. 10 shows another example of a computing system showing external data sources and client data sources that ingested by a more general Al computing environment and a client Al computing environment.

[0138] In some cases, the client Al agent outputs one or more of: a score, a SWOT analysis, a STAR report, an executive summary with an explanation for a score, an explanation for the SWOT analysis, and a recommendation. In some cases, the explanation can include citations to actual structured data (transactions), or citations to unstructured data (contracts, business rules, policies, and procedures, etc.), or both.

[0139] In FIG. 10, a computing system 1000 shows external data sources 1002 and client data sources 1004 that provide data that is ingested by another embodiment of an Al computing environment 1010. In some cases, the Al computing environment 1010 executes the operations described with respect to the Al computing environment 110, including computations of an agentic Al compute environment 118.Atty. Dkt. No. 142937-5001

[0140] In some cases, the Al computing environment includes one or more client Al agents 1012, which are part of one or more data pipelines. The one or more data pipelines include: pattern recognition and anomaly detection, Al models, vendor insights, a labeling module, training data, a predictive analytics module, a risk assessment module, a compliance module, a vendor vector store, an ontology / taxonomy of keywords and concepts, or a module for adjustable self-tuning of weights and bias, or one or more various combinations thereof.

[0141] In some cases, a data pipeline includes both a forecast module configured to generate forecasts and a promotion module configured to suggest autonomous product, price, place, and promotion (4Ps) recommendations. In some cases, the promotion module autonomously analyses firmographic, demographic, and exogenous data, and corresponding forecasted data from the forecast module, to generate the 4Ps recommendations. In some cases, the forecast module computes forecasts for stock and the promotion module uses the forecasts to generate a new old stock (NOS) autonomous recommendation to sell inventory that is new but old and apply the 4Ps to sell and get the inventory off the books. In some cases, the forecast module includes a look-a-like I cosine similarity sub-module to model and forecast new products using the 4Ps recommendations to help model and forecast new product introductions based on prior customer, vendor, firmographic and demographic similar historical structured and unstructured data and machine learnings.

[0142] In some cases, the client Al agent 1012 generates outputs 1020, including: a score, a Strength Weakness Opportunities Threats (SWOT) analysis, an explanation for the score, an explanation for the SWOT analysis, or a recommendation, or one or more various combinations thereof. In some cases, one or some of these outputs 1020 are ingested by the client Al agent 1012 for additional processing.

[0143] FIG. 11 shows another data architecture. In some cases, the computing system 1100 executes the operations described with respect to the Al computing environment 110, including computations of an agentic AI compute environment 118. The computing system 1100, in some cases, is a stack of module that includes underlying infrastructure 1102, data storage 1104, core 1106 (e.g., data pipelines), multiple client Al agents 1108 (e.g., across retail industry, logistics industry, cybersecurity industry, etc.), and a user layer 1110 (e.g., including users interactingAtty. Dkt. No. 142937-5001 directly with client Al agents, or an Al agent interfacing with a client Al agent, or an application programming interface (API) interfacing with a client Al agent, or a combination thereof).

[0144] In some cases, a computing system is provided for training and deploying artificial intelligence (Al) models, the computing system comprising a processor, a communication interface, and memory, the processor coupled to the communication interface and the memory. The memory comprises a fundamental Al model that is trained. The processor configured to at least: ingest, from an external data source, industry-specific data; re-train a fundamental Al model using the industryspecific data to generate an industry-specific Al model; deploy the industry-specific Al model; ingest, from a client data source, client data; re-train the industry-specific Al model using the client data to generate a client Al model; deploy the client Al model for access by the controller agentic Al agent or the one or more supporting Al agent, or both; ingest, from the client data source, new client data; re-train the client Al model using the new client data to generate a new client Al model, and deploy the new client Al model in place of the client Al model for access by the controller agentic Al agent or the one or more supporting Al agent, or both.

[0145] In some cases, the computing system further comprises a client Al computing environment and an Al computing environment separate from the client Al computing environment; wherein the industry-specific Al model is deployed within the client Al computing environment; wherein the client Al model is generated and deployed within the client Al computing environment; wherein the new client Al model is generated and deployed within the client Al computing environment; and wherein the industry-specific Al model is generated in the Al computing environment.

[0146] In some cases, the client Al computing environment comprises one or more security permissions to correspond with the client data source; and wherein and the one or more security permissions, the client data source, the client Al model, and the new client Al model are inaccessible by the client Al computing environment.

[0147] In some cases, the computing system further comprises a second client Al computing environment that is separate from the client Al computing environment and the Al computing environment. The processor is further configured to: deploy the industry-specific Al model in the second client Al computing environment; ingest, fromAtty. Dkt. No. 142937-5001 a second client data source accessible by the second client Al computing environment, second client data; re-train the industry-specific Al model using the second client data to generate a second client Al model; deploy the second client Al model in the second client Al computing environment; ingest, form the second client data source, new second client data; re-train the second client Al model using the new second client data to generate a new second client Al model, and deploy the new second client Al model in place of the second client Al model in the second client Al computing environment.

[0148] In some cases, the client data source is part of a client system, and the client Al computing environment comprises a digital container that establishes secure communication with the client system.

[0149] In some cases, the processor is further configured to: after deploying the new client Al model, ingest, from the external data source, new industry-specific data; re-train the fundamental Al model or the industry-specific Al model, or both, using the using the new industry-specific data to generate a new industry-specific Al model; deploy the new industry-specific Al model; re-train the new industry-specific Al model using the new client data to generate a second new client Al model; and deploy the second new client Al model in place of the new client Al model.

[0150] In some cases, ingesting the client data comprises: the computing system communicating with multiple data nodes of the client data source; wherein the multiple data nodes comprises: a first data node that generates a first output; a second data node that is configured to process at least the first output to generate a second output; and a third data node that is configured to process at least the second output to generate a third output; and, wherein the first output, the second output and the third output form at least part of the client data.

[0151] In some cases, ingesting the client data comprises: the computing system communicating with multiple data nodes of the client data source; and wherein one or more of the multiple data nodes comprises a node Al agent that is configured to generate node data that forms at least part of the client data.

[0152] In some cases, the multiple data nodes comprise multiple robots.

[0153] In some cases, the multiple data nodes comprise multiple vehicles.

[0154] In some cases, the multiple data nodes comprise multiple machines.Atty. Dkt. No. 142937-5001

[0155] In some cases, the new client Al model communicates with a client computing device, and, when the new client Al model is deployed, the new client Al model is configured to generate a message that is transmitted to the client computing device; and, wherein the message comprises an analysis of at least the client data and the new client data.

[0156] In some cases, the message is automatically generated in response to the computing system detecting a query from the client computing system.

[0157] In some cases, the message is automatically generated in response to the computing system detecting new industry specific data that differs from the industry-specific data.

[0158] In some cases, the message is automatically generated in response to the computing system detecting additional new client data that differs from the new client data and the client data.

[0159] In some cases, the processor is further configured to at least: ingest control data comprising natural language; process the control data using the client Al model or the new client Al model to generate a plurality of natural language logic conditions; when the client Al model or the new client Al model is deployed, receive the client data from a data node that is part of the client data source; determine, using the client Al model, a node data score relevant to one or more of the natural language logic conditions; and, determine if the node data score triggers an urgent condition, and, responsive to triggering the urgent condition, generating and sending a message comprising information about the data node.

[0160] In some cases, the message comprises a natural language description regarding the data node, contextual information obtained from the control data, and compliance or non-compliance based on the one or more of the natural language logic conditions.

[0161] In some cases, when ingesting the client data from the client data source, the processor is configured to provide a graphical user interface comprising a drag-and-drop data collection object; wherein the drag-and-drop data collection object is configured to receive one or more data files via a drag-and-drop interaction, the one or more data files comprising the client data.Atty. Dkt. No. 142937-5001

[0162] Referring to FIG. 12, there is illustrated a simplified block diagram of a computer in accordance with at least some embodiments. Computer 1200 is an example implementation of a computer for the Al computing environment 110, the client Al computing environment 818, and for the other computing devices and servers. Computer 1200 has at least one processor 1210 operatively coupled to at least one memory 1220, at least one communications interface 1230 (also herein called a network interface), and at least one input / output device 1240.

[0163] The at least one memory 1220 includes a volatile memory that stores instructions executed or executable by processor 1210, and input and output data used or generated during execution of the instructions. Memory 1220 may also include non-volatile memory used to store input and / or output data - e.g., within a database - along with program code containing executable instructions.

[0164] Processor 1210 may transmit or receive data via communications interface 1030, and may also transmit or receive data via any additional input / output device 440 as appropriate.

[0165] In some cases, the processor 1210 includes a system of central processing units (CPUs) 1212. In some other cases, the processor includes a system of one or more CPUs and one or more Graphical Processing Units (GPUs) 1214 that are coupled together. Although not shown, in combination or in alternative to GPUs, the CPUs are also coupled with tensor processing unit (TPUs), and / or other Al- dedicated processors. In some cases, the GPUs, TPUs and / or other Al-dedicated processors execute machine learning computations or neural network computations, or both. In some cases, the GPUs, TPUs and / or other Al-dedicated processors execute the operations of the Al agents described herein.

[0166] The aforementioned data store and computing could be at one business or N number of businesses within an ecosystem of businesses.

[0167] A combination of the aforementioned methods could be used to orchestrate agentic agents throughout an industry and business ecosystem.

[0168] Various systems or processes have been described to provide examples of embodiments of the claimed subject matter. No such example embodiment described limits any claim and any claim may cover processes or systems that differ from those described. The claims are not limited to systems or processes having allAtty. Dkt. No. 142937-5001 the features of any one system or process described above or to features common to multiple or all the systems or processes described above. It is possible that a system or process described above is not an embodiment of any exclusive right granted by issuance of this patent application. Any subject matter described above and for which an exclusive right is not granted by issuance of this patent application may be the subject matter of another protective instrument, for example, a continuing patent application, and the applicants, inventors or owners do not intend to abandon, disclaim or dedicate to the public any such subject matter by its disclosure in this document.

[0169] For simplicity and clarity of illustration, reference numerals may be repeated among the figures to indicate corresponding or analogous elements. In addition, numerous specific details are set forth to provide a thorough understanding of the subject matter described herein. However, it will be understood by those of ordinary skill in the art that the subject matter described herein may be practiced without these specific details. In other instances, well-known methods, procedures, and components have not been described in detail so as not to obscure the subject matter described herein.

[0170] The terms “coupled” or “coupling” as used herein can have several different meanings depending in the context in which these terms are used. For example, the terms coupled or coupling can have a mechanical, electrical or communicative connotation. For example, as used herein, the terms coupled or coupling can indicate that two elements or devices are directly connected to one another or connected to one another through one or more intermediate elements or devices via an electrical element, electrical signal, or a mechanical element depending on the particular context. Furthermore, the term “operatively coupled” may be used to indicate that an element or device can electrically, optically, or wirelessly send data to another element or device as well as receive data from another element or device.

[0171] As used herein, the wording “and / or” is intended to represent an inclusive-or. That is, “X and / or Y” is intended to mean X or Y or both, for example. As a further example, “X, Y, and / or Z” is intended to mean X or Y or Z or any combination thereof.

[0172] Terms of degree such as "substantially", "about", and "approximately" as used herein mean a reasonable amount of deviation of the modified term such thatAtty. Dkt. No. 142937-5001 the result is not significantly changed. These terms of degree may also be construed as including a deviation of the modified term if this deviation would not negate the meaning of the term it modifies.

[0173] Any recitation of numerical ranges by endpoints herein includes all numbers and fractions subsumed within that range (e.g., 1 to 5 includes 1 , 1.5, 2, 2.75, 3, 3.90, 4, and 5). It is also to be understood that all numbers and fractions thereof are presumed to be modified by the term "about" which means a variation of up to a certain amount of the number to which reference is being made if the result is not significantly changed.

[0174] Some elements herein may be identified by a part number, which is composed of a base number followed by an alphabetical or subscript-numerical suffix (e.g., 312a, or 312b). All elements with a common base number may be referred to collectively or generically using the base number without a suffix (e.g., 312).

[0175] The systems and methods described herein may be implemented as a combination of hardware or software. In some cases, the systems and methods described herein may be implemented, at least in part, by using one or more computer programs, executing on one or more programmable devices including at least one processing element, and a data storage element (including volatile and non-volatile memory and / or storage elements). These systems may also have at least one input device (e.g. a pushbutton keyboard, mouse, a touchscreen, and the like), and at least one output device (e.g. a display screen, a printer, a wireless radio, and the like) depending on the nature of the device. Further, in some examples, one or more of the systems and methods described herein may be implemented in or as part of a distributed or cloud-based computing system having multiple computing components distributed across a computing network. For example, the distributed or cloud-based computing system may correspond to a private distributed or cloud-based computing cluster that is associated with an organization. Additionally, or alternatively, the distributed or cloud-based computing system be a publicly accessible, distributed or cloud-based computing cluster, such as a computing cluster maintained by Microsoft Azure™, Amazon Web Services™, Google Cloud™, or another third-party provider. In some instances, the distributed computing components of the distributed or cloudbased computing system may be configured to implement one or more parallelized, fault-tolerant distributed computing and analytical processes, such as processesAtty. Dkt. No. 142937-5001 provisioned by an Apache Spark™ distributed, cluster-computing framework or a Databricks™ analytical platform. Further, and in addition to the CPUs described herein, the distributed computing components may also include one or more graphics processing units (GPUs) capable of processing thousands of operations (e.g., vector operations) in a single clock cycle, and additionally, or alternatively, one or more tensor processing units (TPUs) capable of processing hundreds of thousands of operations (e.g., matrix operations) in a single clock cycle.

[0176] Some elements that are used to implement at least part of the systems, methods, and devices described herein may be implemented via software that is written in a high-level procedural language such as object-oriented programming language. Accordingly, the program code may be written in any suitable programming language such as Python or Java, for example. Alternatively, or in addition thereto, some of these elements implemented via software may be written in assembly language, machine language or firmware as needed. In either case, the language may be a compiled or interpreted language.

[0177] At least some of these software programs may be stored on a storage media (e.g., a computer readable medium such as, but not limited to, read-only memory, magnetic disk, optical disc) or a device that is readable by a general or special purpose programmable device. The software program code, when read by the programmable device, configures the programmable device to operate in a new, specific, and predefined manner to perform at least one of the methods described herein.

[0178] Furthermore, at least some of the programs associated with the systems and methods described herein may be capable of being distributed in a computer program product including a computer readable medium that bears computer usable instructions for one or more processors. The medium may be provided in various forms, including non-transitory forms such as, but not limited to, one or more diskettes, compact disks, tapes, chips, and magnetic and electronic storage. Alternatively, the medium may be transitory in nature such as, but not limited to, wire-line transmissions, satellite transmissions, internet transmissions (e.g., downloads), media, digital and analog signals, and the like. The computer usable instructions may also be in various formats, including compiled and non-compiled code.Atty. Dkt. No. 142937-5001

[0179] While the above description provides examples of one or more processes or systems, it will be appreciated that other processes or systems may be within the scope of the accompanying claims.

[0180] To the extent any amendments, characterizations, or other assertions previously made (in this or in any related patent applications or patents, including any parent, sibling, or child) with respect to any art, prior or otherwise, could be construed as a disclaimer of any subject matter supported by the present disclosure of this application, Applicant hereby rescinds and retracts such disclaimer. Applicant also respectfully submits that any prior art previously considered in any related patent applications or patents, including any parent, sibling, or child, may need to be revisited.

Claims

Atty. Dkt. No. 142937-5001What is claimed is:1 . A computing system for controlling artificial intelligence (Al) agents, the computing system comprising: a processor, a communication interface, and memory, the processor coupled to the communication interface and the memory; a controller agentic Al agent operable by the processor, and the controller agentic Al agent communicates with one or more supporting agentic Al agents; and the processor configured to generate a graphical user interface (GUI) comprising a control element that varies a value, and the controller agentic Al agent processes the value to generate one or more controller outputs to control the one or more supporting agentic Al agents.

2. The computing system of claim 1 , wherein the processor is further configured to at least: detect a change of the value of the control element and, in response thereof, input the value to the controller agentic Al agent; autonomously select, by the controller agentic Al agent, one or more supporting Al agents based on processing the value; autonomously generate and send, by the controller agentic Al agent, one or more controlling outputs to each of the one or more supporting Al agents; and update the GUI to display one or more supporting outputs generated by the one or more supporting Al agents.

3. The computing system of claim 2, wherein the controller agentic Al agent receives the one or more supporting outputs from the one or more supporting Al agent; the controller agentic Al agent processes the one or more supporting outputs to generate a controller output; and the processor displays the controller output in the GUI.

4. The computing system of claim 1 , wherein the GUI comprises a slider bar element that displays a current value visual element corresponding to a current value, and displays a target value visual element corresponding to a target value; wherein the control element is the target value visual element and the value is the target value;Atty. Dkt. No. 142937-5001 wherein the target value visual element is operably movable along the slider bar element to adjust the target value.

5. The computing system of claim 1 , wherein the GUI comprises: a set of elements positioned in relation to each other according to a given relationship; a current value visual indicator identifying a current value associated with the set of elements; and a target value visual indicator that identifies a target value associated with the set of elements; wherein the control element comprises a boundary edge corresponding to the target value visual indicator, and the value is the target value; and, the boundary edge is operably movable to bound a different subset of the set of elements and, in response, vary the target value.

6. The computing system of claim 5, wherein the boundary edge is a digitally visualized lariat operably movable by a user input.

7. The computing system of claim 5, wherein the set of elements are positioned in relation to each other according to geographical location.

8. The computing system of claim 5, wherein multiple edges or links are displayed with the set of elements to generate a network graph, wherein the edges or links visually indicate the given relationship.

9. The computing system of claim 1 , wherein the processor is further configured to: detect the control element is in in a first state or position in the GUI corresponding to a first instance of the value; subsequently detect an on-focus state of the control element and, responsive to detecting an off-focus state of the control element following the on-focus state, detect that the control element is a second state or position in the GUI; determine a second instance of the value based on the second state or position in the GUI; and processes the second instance of the value to generate the one or more controller outputs to control the one or more supporting agentic Al agents.Atty. Dkt. No. 142937-500110. The computing system of claim 9, wherein the control element comprises a target value visual element that operably moveable along a slider bar element in the GUI.11 . The computing system of claim 9, wherein the control element comprises a digitally visualized lariat operably movable to highlight a subset of elements from amongst a set of elements displayed in the GUI, wherein the set of elements are displayed positioned in relation to each other according to a given relationship.

12. The computing system of claim 1 , wherein the controller agentic Al agent or the one or more supporting Al agent, or both, ingest client data from multiple data nodes; wherein the multiple data nodes comprises: a first data node that generates a first output; a second data node that is configured to process at least the first output to generate a second output; and a third data node that is configured to process at least the second output to generate a third output; and, wherein the first output, the second output and the third output form at least part of the client data.

13. The computing system of claim 1 , wherein the controller agentic Al agent or the one or more supporting Al agent, or both, ingest client data from multiple data nodes; wherein one or more of the multiple data nodes comprises a node Al agent that is configured to generate node data that forms at least part of the client data.

14. The computing system of claim 1 , wherein the controller agentic Al agent or the one or more supporting Al agent, or both, ingest client data from multiple data nodes; wherein the multiple data nodes comprise multiple robots.

15. The computing system of claim 1 , wherein the controller agentic Al agent or the one or more supporting Al agent, or both, ingest client data from multiple data nodes; wherein the multiple data nodes comprise multiple vehicles.

16. The computing system of claim 1 , wherein the controller agentic Al agent or the one or more supporting Al agents, or both, ingest client data from multiple data nodes; wherein the multiple data nodes comprise multiple machines.Atty. Dkt. No. 142937-500117. The computing system of claim 1 , wherein the processor is further configured to: ingest, from an external data source, industry-specific data; re-train a fundamental Al model using the industry-specific data to generate an industry-specific Al model, wherein the fundamental Al model was previously trained using external data sources; deploy the industry-specific Al model; ingest, from a client data source, client data; re-train the industry-specific Al model using the client data to generate a client Al model; deploy the client Al model for access by the controller agentic Al agent or the one or more supporting Al agent, or both; ingest, from the client data source, new client data; re-train the client Al model using the new client data to generate a new client Al model, and deploy the new client Al model in place of the client Al model for access by the controller agentic Al agent or the one or more supporting Al agent, or both.

18. The computing system of claim 17, further comprising a client Al computing environment and an Al computing environment separate from the client Al computing environment; wherein the industry-specific Al model is deployed within the client Al computing environment; wherein the client Al model is generated and deployed within the client Al computing environment; wherein the new client Al model is generated and deployed within the client Al computing environment; and wherein the industry-specific Al model is generated in the Al computing environment.

19. A method for controlling artificial intelligence (Al) agents, the method executed in a computing environment comprising a processor, a communication interface, and memory, the processor coupled to the communication interface and the memory, and the method comprising: establishing a communication between a controller agentic Al agent and one or more supporting agentic Al agents; andAtty. Dkt. No. 142937-5001 generating a graphical user interface (GUI) comprising a control element that varies a value; and processing, using the controller agentic Al agent, the value to generate one or more controller outputs to control the one or more supporting agentic Al agents.

20. A non-transitory computer readable medium storing computer executable instructions which, when executed by at least one computer processor, cause the at least one computer processor to carry out a method for controlling artificial intelligence (Al) agents, the method comprising: establishing a communication between a controller agentic Al agent and one or more supporting agentic Al agents; and generating a graphical user interface (GUI) comprising a control element that varies a value; and processing, using the controller agentic Al agent, the value to generate one or more controller outputs to control the one or more supporting agentic Al agents.

Images