Short answer

Design customer service systems with a clear protocol for escalating interactions from AI to human agents based on the emotional complexity of the customer's needs.

Field: User-Centred Design
Source: European Journal of Innovation Management (2025)
Method: Comparative analysis and framework development
Evidence: Moderate effect

Current generative AI chatbots, while advanced, cannot fully replicate the empathetic capabilities of human agents, necessitating a careful consideration of when to transition customer interactions to a human. This user-centred design research insight is drawn from a 2025 study published in European Journal of Innovation Management. Using Comparative analysis and framework development, researchers explored how this design variable affects real-world outcomes. The key design takeaway: Design customer service systems with a clear protocol for escalating interactions from AI to human agents based on the emotional complexity of the customer's needs.

Study

User-Centred DesignNew This WeekModerate effect

Generative AI Chatbots Struggle to Replicate Human Empathy in Customer Service

European Journal of Innovation Management · 2025

Key Findings

01Generative AI chatbots, including advanced models like ChatGPT, Gemini, and Copilot, do not possess the empathetic intelligence of human agents.
02The concept of Artificial Emotional Awareness (AEA) can characterize the intuitive intelligence of AI in understanding emotions and triggering the Switch Point (SP).
03A complementary role for human intelligence (HI) and AI is proposed, rather than a complete replacement.

Application

Design takeaway

Design customer service systems with a clear protocol for escalating interactions from AI to human agents based on the emotional complexity of the customer's needs.

How to apply

When designing AI-powered customer service interfaces, integrate a mechanism to detect user sentiment and emotional cues, and program a clear escalation path to a human agent when these cues indicate a need for human empathy.

Project actions

01When designing a user interface for an AI chatbot, consider how you will signal to the user when they are interacting with AI versus a human.
02Think about what kinds of emotional cues a user might exhibit that would indicate the need for human intervention.

Method & Evidence

AimTo determine the optimal 'Switch Point' (SP) for transitioning customer interactions from generative AI chatbots to human agents, considering the AI's ability to manage customer emotions.

MethodComparative analysis and framework development

ProcedureGenerative AI chatbots (ChatGPT-3.5, Gemini, Copilot) were evaluated using the Trait Emotional Intelligence Questionnaire Short-Form (TEIQue-SF) to assess their emotional intelligence. A reference framework was developed to illustrate the concept of the Switch Point (SP) and the proposed Artificial Emotional Awareness (AEA).

ContextCustomer service interactions within the framework of Society 5.0, focusing on human-AI collaboration.

Variables

IVType of AI chatbot (ChatGPT-3.5, Gemini, Copilot)

DVEffectiveness in managing customer emotions, determination of the Switch Point (SP)

CVCustomer service context, use of TEIQue-SF

Strengths & Limitations

Strengths

+Introduces the novel concept of Artificial Emotional Awareness (AEA).
+Provides a framework for understanding the human-AI interaction in customer service.

Limitations

The AI models tested are specific examples; future AI advancements may alter these findings. The study's scope is exploratory, and real-world impact needs more investigation.

Reliability & validity

The use of a standardized questionnaire (TEIQue-SF) contributes to reliability. Validity is supported by the conceptual framework and the comparison between AI and human intelligence, though further empirical validation is noted as a limitation.

Think critically

To what extent can AI ever truly replicate human empathy, and what are the ethical implications of designing systems that attempt to do so?

Design Principles

"Human-AI collaboration in customer service should prioritize emotional intelligence, with defined handover points to ensure optimal user experience."

Understanding the limitations of AI in emotional intelligence is crucial for designing effective human-AI collaboration systems. This insight helps designers create customer service workflows that leverage AI for efficiency while preserving the human touch for complex emotional needs, ultimately improving customer satisfaction and preventing disengagement.

What This Means for Your Design

AI chatbots are good at some things, but they aren't as good as people at understanding and responding to emotions. So, we need to figure out when it's best to switch from talking to a chatbot to talking to a real person.

How to use in your project

1.This research can inform the design of your user interface by suggesting the need for a clear 'handover' mechanism from AI to human interaction.
2.You can use the concept of the 'Switch Point' to justify design decisions related to user flow and agent escalation in your design project.

Add to My Project

Quick Cite

Paragraph starter

This research highlights that while generative AI chatbots like ChatGPT, Gemini, and Copilot show advancements in handling customer interactions, they currently lack the empathetic capabilities of human intelligence. The concept of a 'Switch Point' (SP) is crucial for designing effective human-AI collaboration, ensuring that customer interactions are transitioned to human agents when emotional complexity exceeds AI's capacity. This informs the design of user interfaces that facilitate seamless escalation and maintain a high level of user satisfaction.

Source

European Journal of Innovation Management

The impact of new generative AI chatbots on the switch point (SP): toward an artificial emotional awareness (AEA)

journal · 2025

View source

Questions About This Research

What does the research say about generative ai chatbots struggle to replicate human empathy in customer service?: Design customer service systems with a clear protocol for escalating interactions from AI to human agents based on the emotional complexity of the customer's needs. Evidence: European Journal of Innovation Management (2025).
Why does "Generative AI Chatbots Struggle to Replicate Human Empathy in Customer Service" matter for design?: Understanding the limitations of AI in emotional intelligence is crucial for designing effective human-AI collaboration systems. This insight helps designers create customer service workflows that leverage AI for efficiency while preserving the human touch for complex emotional needs, ultimately improving customer satisfaction and preventing disengagement.
How can designers apply this research?: Design customer service systems with a clear protocol for escalating interactions from AI to human agents based on the emotional complexity of the customer's needs.
What were the main findings?: Generative AI chatbots, including advanced models like ChatGPT, Gemini, and Copilot, do not possess the empathetic intelligence of human agents.. The concept of Artificial Emotional Awareness (AEA) can characterize the intuitive intelligence of AI in understanding emotions and triggering the Switch Point (SP).. A complementary role for human intelligence (HI) and AI is proposed, rather than a complete replacement.
What research method was used?: Comparative analysis and framework development.
How strong is the evidence?: Evidence strength is rated Moderate effect, based on a 2025 journal from European Journal of Innovation Management.
What should I do differently in my next project?: When designing AI-powered customer service interfaces, integrate a mechanism to detect user sentiment and emotional cues, and program a clear escalation path to a human agent when these cues indicate a need for human empathy.
What are the limitations?: The study is exploratory and requires further empirical validation. The impact on real-world customer relationship management is yet to be fully explored.