Commercialization of Trustworthy Artificial Intelligence Agents for Effective Information Collection and Personal Insights Discovery

Abstract

From military operations (e.g., command center virtual agents) to commercial applications (e.g., self-driving cars), human-machine teaming becomes increasingly more common. Under the Air Force Office of Basic Research (AFOSR) Trust and Influence Program, our research is on discovering trust antecedents that influence individualsÕ trust in machines to enable more trustworthy and effective human-machine teaming. Based on our research findings, we envision the emergence of a new generation of Artificial Intelligence (AI) agents with three distinct characteristics: (1) they are empathetic because they can automatically infer a userÕs personality traits to understand the user deeply, and (2) they are trustworthy because they can build personalized trust with a user based on their understanding of the user, and (3) they are responsible because they work well with users to accomplish various tasks effectively based on the established human-machine trust. As a start, our initial product will be a special class of such AI agentsÑAI interviewers that exist in the form of a chatbot and can conduct personalized interviews through a text-based conversation. With their three distinct characteristics, not only can our AI interviewers solicit truthful information from a target audience through a personalized conversation, but they also can automatically infer the personality traits of the audience through the interview to help explain the collected information. Our AI interviewers can be used in a wide variety of applications to help organizations collect information from their target audience effectively and understand the personality of their audience. For example, such applications include interviewing customers in market research, interviewing job applicants for human capital management, and interviewing potential witnesses in law enforcement. Compared to human interviewers or static survey tools, our virtual interviewers will offer three benefits: (a) they can scale to collect deeper information from a large audience through a 1-on-1 interactive, personalized conversation; (b) they can collect more truthful information from users because of their non-judgmental nature; and (c) they can render a more objective assessment of a user because they donÕt have personal emotions and biases. Because of the wide applicability and unique value that our AI interviewers can offer across many industries, we believe that they have enormous commercial potential. We thus are eager to leverage the rigor of the I-Corps program to learn and conduct customer discovery. We can then use the discoveries to build a viable product, formulate go-to-market plans, and accelerate the commercialization of our innovations.

Document Details

Document Type
DoD Grant Award
Publication Date
Oct 16, 2018
Source ID
W911NF1710599

Entities

People

  • Michelle Zhou

Organizations

  • Army Contracting Command
  • Office of the Secretary of Defense

Tags

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Defense Technology Research and Development.
  • Psychometric Testing or Psychological Assessment.

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy