Skip to main content Skip to navigation

How Managers Perceive AI-Assisted Conversational Training for Workplace Communication

Project Overview

The document explores the transformative role of generative AI in education, specifically through the implementation of a tool called CommCoach, aimed at enhancing workplace communication training for managers. By utilizing large language models (LLMs), CommCoach simulates conversational agents that engage users in role-playing scenarios, offering personalized, iterative, and context-aware feedback to improve their communication skills in a safe, low-risk environment. This system encourages reflective learning and deeper comprehension of effective communication strategies through features that support scenario creation, feedback generation, and dialogue management. However, the document also acknowledges significant challenges, including issues of adaptability, potential bias in AI responses, and the necessity for human oversight to ensure effective learning outcomes. Overall, the findings suggest that while generative AI tools like CommCoach hold substantial promise for developing communication competencies in educational settings, careful consideration of their limitations is essential for successful implementation.

Key Applications

CommCoach - AI-assisted communication training system

Context: Workplace communication training for managers and employees, focusing on enhancing communication skills through role-playing scenarios and real-time feedback.

Implementation: Developed as a web-based application using GPT-4-powered agents in a client-server architecture. It features a user-friendly interface that mimics popular chat applications, allowing users to engage in role-playing scenarios and receive feedback on their communication skills.

Outcomes: ['Improved communication skills', 'Increased user engagement', 'Real-time feedback to enhance learning', 'Ability to explore multiple conversational paths']

Challenges: ['Potential biases in AI responses', 'User resistance to AI feedback', 'Balancing adaptability with consistency in feedback', 'Managing potential biases in AI-generated personas', 'Need for continuous updates to scenario relevance']

Implementation Barriers

Technological/Ethical

AI lacks the contextual understanding that human mentors provide, which is essential for effective training. There is also potential misuse of AI for performance evaluation rather than developmental training, compromising trust and autonomy.

Proposed Solutions: Integrate human expertise into AI training processes to contextualize feedback and enhance training effectiveness. Implement transparency and ethical guidelines around the use of AI in workplace training.

Technical Barrier

Challenges related to ensuring the AI's responses are unbiased and appropriate, as well as maintaining the system's reliability and accuracy.

Proposed Solutions: Implement thorough testing and evaluation protocols to minimize bias and improve response quality.

User Acceptance Barrier

Resistance from users to accept AI-generated feedback or guidance over traditional methods.

Proposed Solutions: Providing clear explanations of the AI's benefits, training users on how to effectively interact with the system.

Project Team

Lance T. Wilhelm

Researcher

Xiaohan Ding

Researcher

Kirk McInnis Knutsen

Researcher

Buse Carik

Researcher

Eugenia H. Rho

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Lance T. Wilhelm, Xiaohan Ding, Kirk McInnis Knutsen, Buse Carik, Eugenia H. Rho

Source Publication: View Original PaperLink opens in a new window

Project Contact: Dr. Jianhua Yang

LLM Model Version: gpt-4o-mini-2024-07-18

Analysis Provider: Openai

Let us know you agree to cookies