Pindrop Launches BotStopper to Identify AI Voice Agents in Real Time

Pindrop has launched BotStopper, a real-time detection system designed to help businesses distinguish AI voice agents and automated callers from humans.

The technology arrives as autonomous agents increasingly contact banks, healthcare providers, insurers and retailers on behalf of consumers. The emerging challenge for contact centers is not simply blocking bots, but determining which automated callers are legitimate, who authorized them and what systems they should be permitted to access.

BotStopper uses Pindrop Pulse, the company’s deepfake detection engine, to classify a caller as human or machine in roughly two seconds. It is available as a standalone product, allowing organizations to deploy it without purchasing Pindrop’s broader identity security platform.

The system is supported by Pindrop’s new AI Voice Consortium, which launches with a database of more than 5,000 AI-generated voices. That registry is intended to help enterprises recognize known voice agents instead of receiving only a generic warning that audio may be synthetic.

“The goal is not to block AI agents. Legitimate agents are going to call enterprises on behalf of real customers and they should be welcomed,” said Vijay Balasubramaniyan, CEO and co-founder of Pindrop. “The goal is to know which agent is calling, who it represents and what it should be allowed to do.”

Pindrop said AI-powered attacks increased nearly 1,680 percent since late 2024 across more than 700 million customer interactions. The company has observed automated systems impersonating customers, probing contact-center processes and consuming resources while adapting their responses when challenged.

In one Fortune 500 healthcare deployment, the organization reportedly detected more than 30,000 bot calls in less than a year. Bot activity dropped 94.3 percent within four months of introducing Pindrop’s real-time detection technology. At its peak, the operation recorded 6,668 bot calls in one month while approximately $15.1 million in health savings account balances was potentially exposed.

Optum Financial, part of UnitedHealth Group, is also using Pindrop Pulse to identify suspicious automated activity before calls reach customer service representatives.

BotStopper represents an early attempt to build an identity layer for AI-to-business communication. As autonomous agents begin filing claims, checking account balances and conducting transactions, enterprises will need policies that treat machine identity as a distinct security problem rather than another form of robocalling.

Pindrop is also expanding its deepfake detection tools into virtual meetings and hiring platforms, targeting proxy candidates, identity switching and synthetic participants during remote interviews.