The Case Against Using AI in Customer-Facing Roles

Hi! I’m Abbey, Head of Marketing at Gnatta. We recently made the decision to ask key members of the Gnatta team to make a direct, personal contribution on a thought leadership topic of their choice. Being Head of Marketing, I carried myself to the altar of sacrifice, and have taken the first contribution! Lead by example, and so on and so forth…

But as many marketers have learned in the past 18 months, we no longer need to write our own content word for word. Instead, we can kick up our feet, and ask ChatGPT to do it for us. Ultimately more articulate and better proofed than most of us ever were in the first place – it’s as good a method as any for getting words on the page. Sometimes.

I have one big gripe with the use of AI. And that gripe, is that it is not human. Here’s why I think that’s a problem.

P.S. My notes in ChatGPT’s response will [look like this].

What I Said To ChatGPT:

See the transcript here. 

Can you help me write a [300-word] commentary on why AI (like yourself) should not be used as a customer-facing tool? Here’s the gist of what I want to say:

Conversational AI is only as good as the data it has access to, and cannot truly empathise like a human being can. Therefore it is limited.

So I can use AI to write this blog (like I am doing – let’s tell the readers that you’re the author!) because I don’t need data to present my opinion that is not based on statistical fact. + a blog doesn’t particularly need empathy, or to make a complex decision!

But, AI is fundamentally limited because it cannot make decisions in the same way a human being can. In Gnatta, we offer our clients an AI Assistant to support human agents by summarising the customer interaction up to that point, or by drafting replies (linguistic prowess is AIs greatest skill!). AI cannot, however, decide if it would be correct in sending that reply – because it can’t ‘read the room’ like a human being can.

As an AI yourself, what do you think? Can you make critical customer care decisions and be 100% certain it’s as good as a decision made by a human agent?

What It Said Back:

In the modern era, AI has become ubiquitous [reason #1 AI shouldn’t be customer facing – the vocabulary!] in streamlining business operations. However, when it comes to customer-facing roles, relying solely on AI may not be ideal. While conversational AI like myself can mimic human interaction to some extent, it lacks the empathy essential for effective customer service.

Gnatta, for instance, uses AI Assistants [an AI Assistant, in the singular] to support human agents by summarizing [should’ve provided a style guide/preference for British English] interactions and drafting responses. Yet, AI cannot read emotions or assess the context of a situation like humans can. In scenarios requiring delicate handling, AI’s responses may lack the nuance and empathy necessary to effectively address customer concerns.

Human agents possess the ability to empathize, understand emotions, and adapt their responses accordingly. They can pick up on subtle cues and adjust their approach to meet the customer’s needs. In contrast, AI operates within the confines of its programming, unable to deviate or make complex decisions beyond its predefined parameters. [In other words, it can empathise only in the way it has been shown how to empathise]. 

Can AI make critical customer care decisions with the same certainty as humans? [Here’s the crux of it!] The answer is no. While AI can handle routine inquiries, it lacks the human touch crucial for navigating the complexities of customer interactions. Businesses risk alienating customers by relying too heavily on AI, as authentic human connections are often what customers seek when reaching out for support. [This, I think, is the ‘take-home’ wisdom].

In conclusion, while AI has its merits, it is not a suitable substitute for human empathy in customer-facing roles. Striking a balance between AI-driven efficiency and the human touch is essential for businesses aiming to deliver exceptional customer experiences.