Monitoring agent oversight of artificial intelligence content in a contact center
Embodiments provide an oversight mechanism using eye tracking to monitor a human agent's review of artificial intelligence generated contact in a contact center communication session. Content generated by an artificial intelligence engine for each session can be presented to the human agent through a user interface for review by the human agent. While the generated content is being presented, an eye tracking system associated can monitor review of the generated content by the human agent. An indication of the human agent's level of attention to the generated content can be generated based on monitoring the review of the content by the human agent. The indication of the human agent's level of attention can be stored in a set of Key Performance Indicators (KPIs) for the human agent. One or more reports can be provided based on the KPIs.
1. A method for monitoring of agent oversight of automatically generated content in a contact center, the method comprising:
conducting, by one or more communication modules of the contact center over one or more communication channels, a communication session with each of a plurality of client devices, each client device associated with a customer contact;
generating, by an Artificial Intelligence (AI) engine of the contact center, AI-generated content for each communication session;
providing, by the AI engine, the AI-generated content for each communication session to an agent device associated with a human agent prior to providing the generated content to any of the client devices, wherein the provided, AI-generated content is presented to the human agent through a user interface provided by the agent device;
monitoring, by an eye tracking system associated with the agent device, review of the AI-generated content by the human agent while the AI-generated content is presented through the user interface provided by the agent device; and
generating, by the eye tracking system associated with the agent device, an indication of the human agent's level of attention to the AI-generated content based on monitoring the review of the generated content by the human agent.
2. The method of claim 1 , further comprising storing the generated indication of the human agent's level of attention in a set of Key Performance Indicators (KPIs) for the human agent.
3. The method 2 , further comprising providing one or more reports based on the KPIs.
4. The method of claim 1 , wherein the communication sessions with the plurality of customer contacts comprise one or more of a chat session, a Short Message Service (SMS) exchange, and email exchange, or another text-based exchange between the customer contacts and the human agent.
5. The method of claim 1 , wherein the generated indication of the human agent's level of attention is relative to a predetermined expected level of attention.
6. The method of claim 5 , wherein generating the content for each communication session further comprises generating a level of confidence for the AI-generated content.
7. The method of claim 6 , wherein the predetermined expected level of attention is based on the generated level of confidence for the AI-generated content.
8. The method of claim 6 , further comprising providing an indication of the generated level of confidence for the AI-generated content through the user interface provided by the agent device and along with the AI-generated content.
9. A system comprising:
a processor; and
a memory coupled with and readable by the processor and storing therein a set of instructions which, when executed by the processor, causes the processor to monitor agent oversight of automatically generated content in a contact center by:
conducting, by one or more communication modules of the contact center over one or more communication channels, a communication session with each of a plurality of client devices, each client device associated with a customer contact;
generating, by an Artificial Intelligence (AI) engine of the contact center, AI-generated content for each communication session;
providing, by the AI engine, the AI-generated content for each communication session to an agent device associated with a human agent prior to providing the AI-generated content to any of the client devices, wherein the provided, AI-generated content is presented to the human agent through a user interface provided by the agent device;
monitoring, by an eye tracking system associated with the agent device, review of the AI-generated content by the human agent while the AI-generated content is presented through the user interface provided by the agent device; and
generating, by the eye tracking system associated with the agent device, an indication of the human agent's level of attention to the AI-generated content based on monitoring the review of the AI-generated content by the human agent.
10. The system of claim 9 , further comprising storing the generated indication of the human agent's level of attention in a set of Key Performance Indicators (KPIs) for the human agent and providing one or more reports based on the KPIs.
11. The system of claim 9 , wherein the communication sessions with the plurality of customer contacts comprise one or more of a chat session, a Short Message Service (SMS) exchange, and email exchange, or another text-based exchange between the customer contacts and the human agent.
12. The system of claim 9 , wherein the generated indication of the human agent's level of attention is relative to a predetermined expected level of attention and wherein generating the content for each communication session further comprises generating a level of confidence for the AI-generated content.
13. The system of claim 12 , wherein the predetermined expected level of attention is based on the generated level of confidence for the AI-generated content.
14. The system of claim 12 , further comprising providing an indication of the generated level of confidence for the AI-generated content through the user interface provided by the agent device and along with the generated content.
15. A non-transitory computer-readable medium comprising a set of instructions stored therein which, when executed by a processor, causes the processor to monitor agent oversight of automatically generated content in a contact center by:
conducting, by one or more communication modules of the contact center over one or more communication channels, a communication session with each of a plurality of client devices, each client device associated with a customer contact;
generating, by an Artificial Intelligence (AI) engine of the contact center, AI-generated content for each communication session;
providing, by the AI engine, the AI-generated content for each communication session to an agent device associated with a human agent prior to providing the AI-generated content to any of the client devices, wherein the provided, AI-generated content is presented to the human agent through a user interface provided by the agent device;
monitoring, by an eye tracking system associated with the agent device, review of the AI-generated content by the human agent while the AI-generated content is presented through the user interface provided by the agent device; and
generating, by the eye tracking system associated with the agent device, an indication of the human agent's level of attention to the AI-generated content based on monitoring the review of the AI-generated content by the human agent.
16. The non-transitory computer-readable medium of claim 15 , further comprising storing the generated indication of the human agent's level of attention in a set of Key Performance Indicators (KPIs) for the human agent and providing one or more reports based on the KPIs.
17. The non-transitory computer-readable medium of claim 15 , wherein the communication sessions with the plurality of customer contacts comprise one or more of a chat session, a Short Message Service (SMS) exchange, and email exchange, or another text-based exchange between the customer contacts and the human agent.
18. The non-transitory computer-readable medium of claim 16 , wherein the generated indication of the human agent's level of attention is relative to a predetermined expected level of attention and wherein generating the content for each communication session further comprises generating a level of confidence for the AI-generated content.
19. The non-transitory computer-readable medium of claim 18 , wherein the predetermined expected level of attention is based on the generated level of confidence for the AI-generated content.
20. The non-transitory computer-readable medium of claim 18 , further comprising providing an indication of the generated level of confidence for the AI-generated content through the user interface provided by the agent device and along with the AI-generated content.