The GenAI hype is over — now comes the hard part

The generative AI reality check hit us like a ton of bricks in slow motion. One minute, it was being hailed as the ultimate solution for every customer experience challenge under the sun. The next, we were left wondering why the transformation hadn’t materialized quite the way we expected.

I spoke with Gillian Heltai, a customer revenue and services leader with over 15 years of experience scaling B2B companies like Lattice and Talkdesk. As an Operator in Residence at Emergence Capital, she works closely with founders on customer retention and growth strategies. Her perspective on generative AI cuts through the hype, offering a grounded view of what it really takes to transform CX operations.

The AI “revolution” was never real

“AI transformation” has become the headline of the year. But for those in the thick of customer experience operations, the reality of generative AI feels far more like a gradual evolution than a sweeping revolution.

"One of the big learnings through this process of evolving our businesses is that it’s not moving as fast as people feel or fear it is," Gillian explained.

There have been wins — automating routine tasks, personalizing interactions at scale, and delivering quicker resolutions. But calling this a transformation? That’s a stretch.

"If I’m honest, these were things we probably shouldn’t have had humans doing anyway,” Gillian noted. “It’s still a lot of low-hanging fruit that could’ve been solved without GenAI, using other AI, ML, or NLP methods."

Generative AI has added value, no doubt. But deploying AI doesn’t automatically translate into a fully transformed operation. In fact, 63% of CX leaders report that the financial investment required to implement AI has been higher than expected, underscoring the gap between expectations and reality​.

That’s not to say transformation isn’t possible (it absolutely is). But real transformation? "I don’t know if the time horizon is two years or 10 years,” Gillian said. “But there will be a point where we look back and think, ‘I can’t believe we used to do it that way.’"

Real change takes time and intention. You need the right strategy, the right infrastructure, and the right people in place to truly harness what AI can do. And that doesn’t happen with the flip of a switch.

The long game: build for the future, not for flash

Real transformation in customer experience isn’t about flashy AI demos or quick wins — it’s about playing the long game. The companies that see lasting, meaningful change know that AI isn’t a magic wand. Instead, they view it as a tool in a broader strategy, one that requires careful planning and a foundation built for scalability.

"The timing of GenAI was perfect — right when everyone was trying to figure out efficiency,” Gillian reflected. “It was like, 'Oh, look! A silver bullet.' But there never is a silver bullet.”

To make AI effective, you first need a rock-solid knowledge base. AI is only as smart as the data it’s trained on, so if your knowledge base is outdated or incomplete, even the most advanced AI will deliver subpar results. Strong data is essential to prevent costly mistakes and ensure your AI delivers accurate, useful information.

Next, AI needs to be integrated seamlessly across your tech stack — from your CRM to communication channels to agent workflows. Without strong connections between these systems, your AI won’t reach its full potential, leading to inconsistent results and customer frustration.

Finally, transformation hinges on how AI fits into your team’s workflow. AI excels at routine tasks, but the human touch remains critical for more complex, nuanced interactions. Designing workflows with smooth, efficient handoffs between AI and agents allows your team to step in seamlessly when needed. The magic happens when AI and agents work in harmony, not in isolation.

This takes time. There’s no shortcut. Rushing to deploy AI without addressing these core areas can lead to frustration, inefficiency, and even customer dissatisfaction. True transformation is built step by step — so take your time and get it right.

Invest in people, security, and rigorous testing

Generative AI is still in its early stages, and it’s changing fast — and so should your team. Real transformation isn’t just about having the latest tech. It’s about staying ahead and making sure your team grows alongside the tools you use. The best companies don’t just invest in AI; they invest in their people, processes, and security.

Start by investing in upskilling. You can’t just add AI to your operation and expect magic. Your team needs the skills to use this technology effectively. That requires ongoing training, collaboration across departments, and ensuring that your agents feel confident working alongside AI. After all, a well-trained agent equipped with the right tools remains your greatest asset. Still, nearly 48% of service professionals fear they could lose their jobs if they don’t adapt to using AI. Yet, for those who are already working with it, 9 out of 10 report that AI helps them serve customers more efficiently.

"History shows that most roles evolve to work with new technology,” Gillian noted. “I don’t think the role will disappear, but there’s still some fear.”

Next, prioritize security. With AI handling sensitive customer data, the stakes are higher than ever. Data breaches, privacy laws, and security risks are big challenges when using new tech. You must put security first at every step of your AI rollout. That means having strong data rules, making sure your AI tools protect privacy, and always staying ahead of risks. In customer service, trust is key — your customers need to know their data is safe with you.

Finally, be tough when judging AI. "It’s the same as any other tech buy — do your homework. Test the product as much as you can, ask others who have used it, and find customers who weren’t listed as references," Gillian advised. This level of care will help you avoid buying tools that don’t give long-term value.

Be realistic, not jaded

The excitement around generative AI was real, but so are the challenges. The reality is that AI won’t fix everything overnight, and it certainly won’t transform your CX operations unless you’re willing to put in the work. But that’s not a bad thing. The long game is where true, lasting transformation happens. It’s where you build stronger workflows, upskill your team, and put technology in place that works in tandem with your agents.

So, if you’re feeling disillusioned by the lack of instant transformation, know you’re not alone. Real progress takes time, strategy, and intentional effort. But with the right approach, generative AI can be a powerful tool in your CX warehouse. The transformation may not be flashy, but it will be meaningful — and that’s what counts.

‍