ITTech Pulse Exclusive Interview Dr. Vijay Balasubramaniyan, Chief Executive Officer, Pindrop

Dr. Vijay Balasubramaniyan, CEO and Co-Founder of Pindrop, discusses AI-driven identity assurance, deepfake detection, and securing trust in enterprise digital collaboration.

Can you share a brief overview of your journey and what inspired you to become the CEO and Co-Founder of Pindrop?

Pindrop really started with my graduate research. I was focused on identifying features in communication systems that could help characterize wanted and unwanted interactions. It became clear early on that trust is the backbone of any communication system. If you can’t verify who’s on the other end, the entire channel becomes vulnerable.

Back then, this was mostly a VoIP security problem, but as communication shifted further online, the threats evolved. What began as detecting fraud and stopping bad actors in the call center naturally expanded into broader identity assurance. In recent years, AI has dramatically accelerated this shift, and with voice cloning and deepfakes rising more than 1,300% year over year, online communication is the new attack surface for bad actors. This evolution is what has led to Pindrop as it is today. The same science-driven approach we used to stop phone fraud now underpins our ability to verify “real humans” and detect synthetic or manipulated content in real time. What started as academic research has become the foundation for how we help major enterprises, governments, and the people they reach navigate an AI-driven world.

What excites you most about Pindrop’s partnership with Cisco Webex?

The most exciting part is the impact we’re able to have. Webex has millions of users and facilitates mission-critical conversations every day for global enterprises through Webex Meetings and Webex Contact Centers. Bringing real-time deepfake detection and participant authentication directly into that environment is a meaningful step forward for digital trust.

Read More: ITTech Pulse Exclusive Interview with Linir Zamir, Team Lead-AI Engineer at Telit Cinterion

What also makes this partnership meaningful is how deeply integrated Pindrop has become within Cisco’s broader ecosystem. Our solutions aren’t sitting on the sidelines; they’re built into the Webex Suite and are now part of how Cisco teams position collaboration security with their customers. That level of alignment allows us to reach enterprises at scale and help them adopt a more proactive approach to identity assurance.

What were the biggest technical challenges in integrating Pindrop’s AI-driven fraud and deepfake detection into the Webex platform?

The biggest challenge was doing this in real time without slowing anything down. These Interactions are unpredictable since people join from different devices, networks, and environments, and deepfake models keep evolving. We had to design a system that could handle all of that: process audio and video continuously, spot subtle signs of manipulation, and respond in milliseconds.

And because Webex operates at a massive global scale, the solution had to be efficient and invisible to users. Balancing speed, accuracy, and reliability was what made the final product feel seamless.

How do you envision the role of multi-modal deepfake detection evolving as enterprise collaboration becomes more AI-driven?

Multi-modal detection is becoming a core part of workplace collaboration. As companies rely more on AI in their day-to-day collaboration, deepfakes won’t appear in just one format — they’ll show up across voice, video, and even behavior. Looking at audio or video in isolation simply isn’t enough anymore.

As AI grows more capable, enterprises will need their own safety systems — AI running quietly in the background that confirms, moment to moment, that the participant is a real human and not something synthetic. Over time, I expect this to become a standard safeguard across industries.

How do you ensure that Pindrop’s security layer scales smoothly across millions of global Webex users?

Pindrop has been operating at a large scale for more than a decade. We already secure billions of interactions every year for major banks, insurers, healthcare organizations, and large telecom providers, so the foundation was well established.

For Webex, we applied the same principles of bringing Pindrop’s enterprise-grade security directly into the collaboration experience, so it works seamlessly in the background. By integrating capabilities like passive voice authentication, advanced fraud detection, and multi-modal deepfake detection, Webex can protect meetings without disrupting how people work. The goal is for the system to operate quietly in the background, across millions of users, without affecting performance.

Read More: ITTech Pulse Exclusive Interview with Michael Campell, Chief Product Officer, Hyland

How do you balance real-time security performance with the need to maintain user experience across large enterprise customers?

Very early on, we saw that many organizations tried to solve security challenges by adding more steps for their users — more questions, more verification, more friction. That approach slows people down, increases operational costs, and still doesn’t stop sophisticated attackers. From the beginning, Pindrop took a different view. We built our platforms to strengthen security while actually reducing friction for genuine users, whether they’re calling into a contact center or joining a meeting. The idea is that the system should protect the meeting without changing the way people work.

This is why our detection runs passively in the background, so legitimate users can authenticate naturally without changing how they interact, while bad actors are identified in real time. That balance is important: enterprises get stronger protection and better user experience at the same time. Enterprise customers care deeply about both security and usability, so we’ve designed the platform to deliver on both.

As a leader, how do you guide teams to innovate quickly while staying ahead of rapidly evolving AI-powered fraud threats?

Pindrop has been a research-driven company from the start. Our R&D and engineering teams work closely together, constantly analyzing new models, studying real attack patterns, and refining our detection methods. That approach has led to more than 300 patents and industry-leading performance in key benchmarks like ASVSpoof and the ACM multimedia deepfake challenge.

My job is to create an environment where we move quickly but stay grounded in real-world data. Innovation is important, but it has to hold up against actual threats. Fraudsters adapt quickly, and we have to stay a step ahead — that mindset is core to how the team operates.

What industry trends do you believe will define the next phase of secure digital collaboration?

The big mega-trend is identity becoming the main attack surface as AI-powered impersonation accelerates and the underlying technology keeps improving. Verifying that every participant is a real human and the right human will be essential. Collaboration platforms will need to build authenticity into their core workflows rather than relying on after-the-fact analysis.

We’ll also see AI agents become active participants in meetings, which means organizations will need clear boundaries between what’s human-generated and what’s AI-generated. And deepfake detection will become a standard, much like spam filtering did for email. Altogether, these trends point to trust and authenticity becoming the bedrock of secure enterprise communication.

How do you see the future of voice security evolving in the next few years?

Voice will continue to be a powerful identity signal, but only if we can reliably distinguish real from synthetic. Over the next few years, I expect to see real-time deepfake detection built directly into communication tools, and voice authentication paired with video, device, behavioral, and location signals for a more complete view of identity.

We’ll also move toward continuous verification, not just checking who someone is at the start of a call, but throughout the interaction. And in regulated sectors especially, deepfake defense will become a standard requirement. Ultimately, voice security will play a central role in establishing trust in digital communication.

Thank you, Dr. Vijay, for taking the time to share your insights with us.

Write to us [⁠wasim.a@demandmediaagency.com] to learn more about our exclusive editorial packages and programmes.