Enterprises are leaving major AI customer experience gains on the table

Enterprises are investing heavily in AI customer experience, but many experiments are stalling out before ever going live.

The culprit is a confidence gap, according to Chris Mina (pictured), chief product and technology officer at LivePerson Inc. Organizations build compelling proof-of-concept workflows and run successful demos, but freeze up when edge cases materialize or legal and security councils voice concerns. That paralysis is keeping enterprise AI customer experience adoption in the single digits, despite steadily increasing consumer expectations for AI-powered service.

“They build these great proof of concepts. They have these workflows — and they sit around,” Mina said. “They run some happy path use cases and say, ‘Wow, this is amazing.’ Then they hit the edge cases and they stop.”

Mina spoke with theCUBE’s John Furrier and co-host Alison Kosik at Google Cloud Next, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed how LivePerson is addressing enterprise AI adoption friction through synthetic testing, real-time orchestration and a deepening partnership with Google Cloud. (* Disclosure below.)

AI customer experience demands confidence before deployment

LivePerson’s answer to the confidence gap is Syntrix, a testing platform that lets brands simulate thousands of customer scenarios using synthetic users and generative test cases before ever deploying a new agent or starting a campaign. The goal is to give brands the documented evidence they need to satisfy internal reviews, according to Mina.

“When [enterprises] get blocked, whether it’s [by] their security, their legal or AI councils, they can come with a set of data that says, ‘Look, we’ve tested in all of these scenarios,'” he said. “‘We feel confident and we’ve got the data and evidence to back it.'”

Beyond testing, LivePerson built a “guardian agent” that monitors 100% of live conversations in real time for both humans and bots, continuously evaluating whether each interaction is on track or requires escalation, Mina explained. The company has also been overhauling its infrastructure, recently completing a multi-year migration to Google Cloud, shedding two decades of on-premises technical debt and providing customers with direct access to Gemini models and hyperscale reliability. With the infrastructure in place, the pressure is now on to meet rising consumer expectations.

“You can’t stop this train as it’s moving forward,” Mina said. “The promise is there to the market. Consumers expect it. We need to help all the brands deliver on this promise safely and securely.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of Google Cloud Next:

(* Disclosure: LivePerson sponsored this segment of theCUBE. Neither LivePerson nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE