Generative AI at Work
Project Overview
The document explores the transformative role of generative AI in education, particularly its applications in enhancing productivity and learning experiences. It highlights that generative AI tools have resulted in a 15% increase in productivity, especially benefiting less experienced and lower-skilled workers in customer support roles. By capturing best practices from high-performing individuals, these AI systems facilitate skill development among lower-skilled agents. The findings indicate improvements in customer sentiment and reductions in worker attrition, reflecting a more positive work environment fostered by AI assistance. However, the document also addresses challenges such as the potential over-reliance on AI technologies and the necessity for continuous training to ensure effective use. Overall, generative AI is positioned as a valuable asset in educational contexts, promoting skill enhancement and improving overall work experiences while highlighting the importance of balancing AI integration with human development needs.
Key Applications
AI assistance for customer support agents
Context: Customer support in a Fortune 500 software company, targeting customer support agents globally, primarily in the Philippines.
Implementation: AI tools were introduced after a randomized pilot program. Agents underwent a three-hour online onboarding session before gaining access to AI recommendations.
Outcomes: Productivity increased by 15%, with notable gains for lower-skilled and less-experienced workers. Improved customer sentiment and reduced requests for manager escalation were observed.
Challenges: AI recommendations sometimes led to reduced quality of conversation for high-skilled workers; reliance on AI could diminish original contributions from experienced agents.
Implementation Barriers
Technical barrier
Generative AI tools may produce misleading information, raising concerns about reliability in high-stakes situations. AI systems can be fine-tuned using feedback from human workers and improved training data.
Proposed Solutions: AI systems can be fine-tuned using feedback from human workers and improved training data.
Organizational barrier
Existing workplace structures may not support the effective implementation of AI tools, requiring complementary organizational investments. Firms need to invest in training and redesign business processes to integrate AI effectively.
Proposed Solutions: Firms need to invest in training and redesign business processes to integrate AI effectively.
Project Team
Erik Brynjolfsson
Researcher
Danielle Li
Researcher
Lindsey Raymond
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Erik Brynjolfsson, Danielle Li, Lindsey Raymond
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