Month after month, AI has woven itself into the fabric of business, transforming daily operations. But what about human connections, and specifically customer relationships? Beyond the grand promises and attractive statistics, what is actually happening on the ground?                                                                        

Valentin Drouet, our Head of Customer Success, Eleazar Baptiste, Head of App Management (RingOver), and Michaël Pudlowski, Digital & Customer Partner (KPMG), explored this topic during a panel discussion at our latest AI Breakfast.

Moving beyond the chatbot cliché

When we talk about AI in customer service, we often immediately picture the classic front-office chatbot that customers interact with to ask questions. 

But what if the most powerful transformation is actually happening in the back office, supporting sales and customer success teams? 

In fact, conversational analysis—AI that listens to, transcribes, and understands customer calls—is rapidly becoming the standard. 

In practical terms: automated CRM entry updates customer records after a call; real-time insights prompt agents with the right rebuttals; and call libraries are used to train new hires using real-world scenarios. 

The result: teams spend less time at their keyboards and more time building relationships.

An example shared by Michaël Pudlowski illustrates this impact: a small online brokerage firm used to employ 10 people just for call monitoring (a regulatory requirement in insurance). With AI, they now monitor 100% of calls instead of 30%, and have redeployed those 10 full-time employees to higher-value tasks.

The customer as an expert

Perhaps one of the biggest impacts of AI on customer relations is that customers now arrive at an agency with a ChatGPT conversation open in another tab. They know—or think they know—their contract better than the advisor. In insurance or banking, that creates a complex dynamic.

Add to this the security and integration constraints for companies that sometimes find themselves with "in-house" AI solutions that are inferior to the consumer tools available to their customers, leaving the door wide open to shadow AI.

AI: a technology rejected by the public?

One might assume that customers insist on speaking with a human advisor, do not trust AI agents, or that using this technology could damage a company's reputation. 

In reality, public opinion appears to be much less polarized than one might imagine.
A recent KPMG customer experience study reveals that what customers want above all else is efficiency, access to an advisor when needed, the fastest possible refunds, and quick answers…

4 out of 5 customers are happy to use a chatbot for simple needs, provided the response is fast and consistent. As soon as complexity increases or the issue becomes critical (telecom outages, insurance claims), the customer wants a human. AI doesn't replace the agent; it redefines their role. 

It is, in fact, simply a means to satisfy a growing need for immediacy.

Welcome to B2A2C

This is arguably one of the most impressive transformations of the coming years. 

Four types of agents now coexist: 

• those deployed by the company
• general-purpose agents like ChatGPT
• browser-integrated agents (Comet, Holo) that perform actions on your behalf
• business-generating agents (Mastercard, Amex, and soon influencers)

The unique feature of the latter two is that they no longer just provide information: they act for you. They navigate, filter, purchase, dispute, and request refunds. The customer still makes the final purchase, but they are no longer the one clicking. This is what we call Business to Agent to Client. 

Within three years, 15% to 30% of customer journeys could be handled by these agents. 

Traditional KPIs—time spent on site, conversion rates, page views—will gradually lose their relevance.

‍

The main pitfalls to avoid for a successful project

Today, there is no shortage of successful use cases and examples of AI's great potential, making it tempting to jump in as quickly as possible. However, be careful not to confuse evolution with haste. Our three panelists discussed the main pitfalls to avoid: 

• Confusing efficiency with vision. Most companies launch AI projects saying, "we're going to save time." That is not a vision; it is a consequence. Michaël Pudlowski cites MAIF, which spent several months co-authoring a charter with its labor unions before launching anything. This groundwork allowed for a much faster deployment later on.
• Forgetting that humans are still "big kids." RingOver experienced this internally: imposing a new tool by directive creates resistance. Explaining what the tool frees up (hated tasks, typing after a call, manual summaries) changes everything. Education is not an "accessory"; it is a prerequisite for the success of any project implementation. 
• Blurring the line between human and AI. Our three speakers were aligned on the need for total transparency. The customer must know who they are talking to and have a choice. Beyond ethics, it is also a strategic calculation: European regulations will mandate this in the coming years, and a "GDPR moment" regarding conversational AI would be very costly.
• Stop thinking of AI as a tool. It is a paradigm shift in how humans and machines divide work within a conversation. Those who treat it as a transformation project rather than a software license purchase will gain a three-year head start.

‍

Final thoughts: Recommendations from each of our speakers

• Michaël Pudlowski (KPMG): "Define your North Star. Have a true long-term vision—12 to 18 months is a long time in AI—and master your data. Without it, there is no AI."
• Eleazar Baptiste (RingOver): "Define the 'why' before the 'how.' Don't just follow trends; look for where the real gains are and keep the customer experience—both internal and external—at the heart of the project."
• Valentin Drouet (Craft AI): "Ask questions and understand the technology. Blindly trusting an AI sold as a magic black box is the worst possible way to get started with AI."

‍

If you would also like to improve efficiency in your customer relations, contact our experts.