AI Boosts Productivity in Customer Service, but Impact Varies Widely Among Workers

In a groundbreaking study published in the Quarterly Journal of Economics, researchers from Stanford University have delved into the impact of generative AI on employee work experiences within the customer service sector. The findings reveal that while less experienced and lower-skilled employees experience significant improvements in both the speed and quality of their work, the opposite is true for highly skilled and professional workers.

The “Generative AI at Work” study was conducted over time at a Fortune 500 firm specializing in business process software. The research team analyzed the use of a new chat assistant by 5,172 customer support agents, revealing that AI assistance increased worker productivity by 15 percent.

Interestingly, the study also found that AI-assisted agents with just two months of tenure performed as well as regular agents with over six months of experience. However, this technology led to a slight decrease in the quality of conversations the most skilled agents conducted.

The researchers also observed that adherence rates to AI recommendations increased over time, and access to AI improved English language fluency, particularly among international agents. Furthermore, the study suggests that using AI improves customer behavior, making interactions less hostile and reducing instances where customers question an agent’s competence or request a supervisor.

In a recent study published by the Quarterly Journal of Economics, researchers have found that artificial intelligence (AI) assistance can significantly improve productivity and work experiences for customer service employees. The effects vary depending on the skill level and knowledge of the workers.

The research was conducted at a Fortune 500 firm that sells business process software, where 5,172 customer support agents used a new chat assistant. The investigation revealed that AI assistance increased worker productivity by 15%, resulting in more chats successfully resolved per hour.

Variability in Impact

The impact of AI assistance on productivity was found to vary significantly among workers. Less-skilled and less experienced employees showed substantial improvements across all productivity measures, with a 30% increase in the number of issues resolved per hour. In contrast, higher-skilled or more experienced workers saw slight gains in speed and minor declines in quality.

The study also found that AI assistance can help workers learn and improve English fluency, particularly for international workers. Newer agents who followed AI recommendations closely generally saw more significant gains in productivity, with adherence rates increasing over time.

Improved Customer Interactions

The researchers also observed that access to AI improved the treatment of agents by customers. Customers were less likely to question agents’ competence or ask to speak to a supervisor, and the tone of customer messages became more positive. This suggests that using AI may improve customer behavior and make contact center work less challenging for employees.

The “Generative AI at Work” study is available on the Oxford University Press website. The findings suggest that AI can be a valuable tool for improving productivity in the customer service sector, particularly for less-skilled and less-experienced workers. However, the impact of AI assistance varies widely depending on the skill level and experience of the employees.

More information
External Link: Click Here For More
Dr. Donovan

Dr. Donovan

Dr. Donovan is a futurist and technology writer covering the quantum revolution. Where classical computers manipulate bits that are either on or off, quantum machines exploit superposition and entanglement to process information in ways that classical physics cannot. Dr. Donovan tracks the full quantum landscape: fault-tolerant computing, photonic and superconducting architectures, post-quantum cryptography, and the geopolitical race between nations and corporations to achieve quantum advantage. The decisions being made now, in research labs and government offices around the world, will determine who controls the most powerful computers ever built.

Latest Posts by Dr. Donovan:

The mind and consciousness explored through cognitive science

Two Clicks Enough for Expert Echolocators to Sense Objects

April 8, 2026
Bloomberg: 21 Factored: Quantum Risk to Crypto Not Imminent Now

Adam Back Says Quantum Risk to Crypto Not Imminent Now

April 8, 2026
Fully programmable quantum computing with trapped-ions

Fully programmable quantum computing with trapped-ions

April 8, 2026