How Consumers Really Feel About AI in Customer Service: Insights from 1,000+ Users Across Six Countries

By: Ian Aitchison

What matters today is not only what AI can do, but how customers respond to it. To understand that shift, COPC Inc. surveyed more than 1,000 consumers across Australia, China, Malaysia, Singapore, the United Kingdom, and the United States. All respondents had interacted with AI-powered customer service within the past three months. 

The results show a CX landscape that is evolving quickly. Customers are willing to use AI, but their trust depends on the resolution, transparency, and how digital and human channels work together. Many organizations have invested heavily in AI, yet our data shows that strong outcomes rely on foundational elements such as journey design, measurement discipline, and operational alignment. 

Consider this common scenario: a customer begins with a chatbot to resolve a billing issue. The bot correctly identifies the account, but it can’t complete the transaction. When the customer is transferred to an agent, the conversation restarts from the beginning. The customer experiences this as poor service. In reality, the issue is not the AI model; it’s how the workflow behind it was designed.

1. AI is Mainstream, but Maturity Varies by Market

Across all six countries, AI has become a normal part of customer service. Chatbots remain the most frequently used AI channel, followed by voice assistants and messaging apps such as WhatsApp, WeChat, and Facebook Messenger. 

Adoption, however, is not uniform. 

• China has near-universal usage, with 99% of consumers reporting an AI interaction in the last three months.
• Australia shows the slowest uptake, with one in four customers having no AI contact at all.

 The United States and Singapore show balanced usage across both chat and voice channels, indicating stronger omnichannel readiness.

These differences illustrate a broader pattern we see across CX programs: technology adoption outpaces operational alignment. Markets that have invested heavily in digital channels often still face challenges with consistency, measurement and design.

2. Satisfaction Rises When Issues Are Resolved

Globally, 74% of customers reported satisfaction with their most recent AI interaction. That number rises sharply when AI fully resolves the issue. Satisfaction soars above 90% when resolution occurs without further steps. When AI fails to resolve the issue, the Net Promoter Score (NPS) can plunge by as much as 70 points. 

This is consistent with decades of COPC research findings. Customers will accept limited empathy or a scripted tone if the interaction is effective. They will not accept unresolved issues or repeated effort.

A practical example:

A U.S. broadband provider implemented an AI assistant to troubleshoot connectivity issues. The AI could identify simple problems, but when escalation occurred, context was not transferred. Resolution rates dropped, and customer effort increased. Once the company redesigned the escalation workflow and ensured that account details, diagnostic steps and prior interactions were visible to the agent, both satisfaction and containment improved.

These improvements were not driven by new technology. They were driven by clear journey design and consistent processes.

3. Transparency Builds Trust

One of the strongest findings in the study is what we refer to as the  ‘transparency dividend.’

Customers who knew they were interacting with AI reported satisfaction rates 34 percentage points higher than customers who were not informed. 

Markets that lead in disclosure, such as Malaysia and Singapore, also lead in overall satisfaction and comfort with AI.

Transparency matters for three reasons:

• It sets expectations about capability.
• It reduces perceived risk.
• It helps customers understand when escalation is appropriate.

We see this consistently in journey mapping work. When organizations provide simple, upfront cues such as “You are speaking with our virtual assistant,” trust improves, and customers judge the interaction more fairly.

4. Handover to Humans Remains the Weakest Link

Across all six markets, the transition from AI to a human agent remains the most frequent point of failure.

• China: 52% of customers reported some form of context loss during escalation.  
• Australia: Only 20% described the handover as seamless. 
• United States: After a failed AI interaction, full resolution rates occurred only about half the time. 

The operational takeaway is clear: AI performance is often judged not by the quality of the algorithm but by how well it connects to human support. In recent COPC assessments, we routinely see five to nine systems involved in a single escalation path. Without a structured design, data is lost, context is dropped, and customers start over. Organizations that perform well in this area typically use service blueprinting and journey-first design. These approaches clarify what information must persist across channels and where accountability sits for each interaction point.

5. Regional Differences Show a Common Pattern

Each country has distinct characteristics, but the overall findings point to a shared operational reality.

• Malaysia: Highest satisfaction (83%) globally, attributed to strong disclosure practices and higher tolerance for digital-first workflows.
• China: High comfort and adoption, but persistent context challenges during escalation. 
• Singapore: Best performance in smooth handovers and multichannel readiness.
• United Kingdom: Moderate satisfaction but weak recovery after failed AI interactions. 
• United States: Highest satisfaction when AI performs well, paired with strong concerns about data privacy and accuracy.
• Australia: Lowest satisfaction overall, driven by higher effort and slower resolution. 

These patterns show that a model or platform does not define the success of AI-enabled CX programs. They are characterized by design, measurement, governance, and integration as discussed in our COPC webinar . 

From Technology to Trust: What Organizations Should Do Next

The research shows that customers are willing to embrace AI in CX, but their trust depends on three things:

• Getting their issue resolved
• Knowing when they’re talking to a machine
• Having a smooth path to a human when they need one

The organizations that get this right design journeys first, then decide where AI should help, how success will be measured, and when a person should take over. 

COPC can help you review your AI-enabled journeys and identify practical changes to strengthen both customer trust and performance.

Learn how to apply these findings to your own CX operation today!