ENABLING A HYBRID SWARM INTELLIGENCE OF HUMANS AND AI AGENTS
Systems and methods are disclosed for enabling a hybrid intelligence of human users and machine agents. In some embodiments, the systems and methods enable a group of humans and AI agents to converge on answers to questions together during a group collaboration session. In some such embodiments, a plurality of collaboration applications communicates with a central server, wherein each collaboration application collects input from and displays output to one of a plurality of collaborating human users. In addition, one or more agent applications are enabled, each agent application controlling at least one machine agent that is configured to participate in the group collaboration session. By receiving and processing user input and agent input during the collaboration, a group response is determined for the hybrid intelligence. In some embodiments, the group response represents a group decision or group forecast and is output to the human users through their collaboration applications.
1 . A hybrid intelligence method for enabling a group of human users and one or more machine agents to converge on an answer to a question during a group collaboration session, the method comprising:
enabling a plurality of collaboration applications to communicate with a central server, each collaboration application configured to collect input from and display output to a respective human user;
enabling one or more agent applications to communicate with the central server, each agent application controlling at least one machine agent;
sending, over a network, a question and a plurality of answer options to each of the collaboration applications and displaying the question and answer options to the respective human users;
capturing, via the collaboration applications, a user intent from each human user, each user intent indicating a preferred answer option and an associated magnitude;
generating, using an artificial intelligence process, an agent intent for each of the one or more machine agents, each agent intent indicating a preferred answer option and an associated magnitude;
receiving, at the central server, the user intents and the agent intents;
processing the user intents and the agent intents to determine a group response for the group of human users and machine agents; and
outputting a representation of the group response to the human users via the collaboration applications, the representation indicating a collaboratively selected answer.
2 . The method of claim 1 , wherein generating the agent intent is performed using an artificial intelligence process modeled on behavior of one or more humans.
3 . The method of claim 1 , wherein the artificial intelligence process generates a probability associated with each of a plurality of answer options and selects the answer option having a highest probability.
4 . The method of claim 1 , wherein each agent application executes on a respective agent computing device in network communication with the central server.
5 . The method of claim 1 , wherein at least one agent application executes on the central server.
6 . The method of claim 1 , wherein each agent intent is generated by simulating a user intent from a human user based on a machine model of human behavior.
7 . The method of claim 1 , wherein the one or more machine agents comprise a plurality of machine agents, each machine agent associated with a different artificial intelligence process.
8 . The method of claim 1 , further comprising displaying a graphical representation of each machine agent to each human user via the collaboration applications.
9 . The method of claim 1 , wherein the group response comprises a textual reply to the question.
10 . The method of claim 1 , wherein the group response comprises a numeric reply to the question.
11 . The method of claim 1 , wherein capturing the user intents, generating the agent intents, and processing the user intents and agent intents are repeatedly performed over a collaboration period to determine a final group response, the final group response being displayed at a conclusion of the collaboration period.
12 . A hybrid intelligence method for enabling a group of human users and one or more machine agents to answer a question during a group collaboration session, the method comprising:
sending a question over a communication network to a plurality of computing devices and causing the question to be displayed to a plurality of human users, each human user associated with a respective computing device;
capturing, via a user interface on each computing device, a user intent from each human user, each user intent indicating a preferred answer to the question;
generating, using an artificial intelligence process, an agent intent for each of the one or more machine agents, each agent intent indicating a preferred answer to the question;
processing the user intents and the agent intents to determine a group response;
repeatedly performing the capturing, generating, and processing over a collaboration period to enable the group of human users and machine agents to converge on a final group response; and
outputting a representation of the final group response to the human users, the representation indicating a collaboratively selected answer.
13 . The method of claim 12 , wherein the artificial intelligence process generates the agent intent by mimicking human behavior.
14 . The method of claim 12 , wherein the artificial intelligence process generates a probability associated with each of a plurality of answer options and selects the answer option having a highest probability.
15 . The method of claim 12 , wherein the group response comprises a textual reply to the question.
16 . The method of claim 12 , wherein the group response comprises a numeric reply to the question.
17 . The method of claim 12 , wherein each machine agent represents a simulated human that expresses intent during the group collaboration session.
18 . The method of claim 1 , further comprising displaying a graphical representation of at least one machine agent on each of the plurality of computing devices