Neural Defend Founder & CEO Piyush Verma on Building the World's Trust Layer for Deepfake Detection

Piyush Verma is the Founder & CEO of Neural Defend (Techstars 2025), a company building a proprietary AI detection foundation model from first principles — not a wrapper on top of someone else's API. Piyush left a research role at MIT to build the company alongside co-founders Siva Shankar and Sumit Singh, with a team based in India.

In June 2026, Neural Defend won the startup award at NYC Tech Week's Everything AI Startup Pitch Competition, taking home roughly $50,000 in equity-free prize money as the only Indian startup among 10 global finalists.

What does your company do, and why is it important?

We're building a proprietary foundation model that detects deepfakes. We don't use open-source algorithms or APIs from companies like Google — what we've built is completely our own technology. Right now we're working with a lot of enterprises and institutions that want to detect deepfakes.

Tell us about your dream client.

I think the dream client would be somebody like an Apple or a Google — companies that could put us in the infrastructure of their products. If Apple created a layer of deepfake detection on their mobile phones, that would be an amazing push for our technology, because it could scale to billions of devices.

You've been described as the "Intel Inside" for deepfake detection. Is that accurate?

Yes, absolutely. We're not just an API or a superficial layer on top — we're in the infrastructure. We're building a trust layer where these companies can include us as part of their core technology.

When did you first realize this was a real, growing problem?

We've been exploring this field for almost a decade now. But I met my co-founders in 2023 — they were already building something in this detection space, and they came to me in late 2023. They wanted support in terms of technology, and they wanted to scale this globally, so we could turn it into a refined product.

How big of a problem is this for global organizations?

The pace at which AI-generated content is growing, in the next couple of years, almost 90% of the content around us will be generated. It's going to be very difficult for us to differentiate what's fake and what's real.

In that context, everything needs a layer of protection in between — something that tells you whether a voice is real or fake, whether an image is real or fake, whether the person on the other side of a payment is real. You can imagine all the payment platforms and communication channels that will need a layer of deepfake detection in the future.

Why did you choose Techstars, and what was the experience like?

Techstars is a global brand — one of the best accelerators we have today, and probably the biggest in terms of the number of companies funded. That brings a lot of brand value, which is very useful to us as an early-stage company.

Techstars has opened several doors for us — a lot of the design partners we're working with right now came through the Techstars network. We connected with some of the portfolio companies, and they introduced us to clients. The mentors, advisors, and managing directors we've met over the last year have been extremely helpful and supportive, and right now Techstars is also helping us find investors for our round. We're very grateful for the opportunity Techstars has given us.

Was there a mentor relationship that stood out?

Techstars connected us with several mentors, especially in the BFSI sector — banking, financial services, and insurance. Some of those mentors introduced us to clients in the Middle East and here in the US. These mentors are actively working with us — it's not transactional, it's a relationship that's grown over time. We're very thankful to Techstars for connecting us with such selfless mentors who are genuinely helping us.

Did the program lead to any pivots for Neural Defend?

One of the core values startups get from Techstars is understanding the customer and the real, burning problem they're facing — and not creating a product that nobody wants. The program was built around solidifying the product and deeply understanding the problem, which is especially important for early-stage founders and startups who haven't found product-market fit yet. Customer discovery — talking to dozens of customers — that's what Techstars helped us with, and it's been a very important part of being in the program.

What would you tell founders who aren't sure an accelerator is right for them?

Don't wait for the perfect opportunity or the right moment. Whenever you're ready to start building, go out into the market and start testing it. If you have a prototype, even if it's not perfect, that's completely fine — go out to the market, talk to customers, get feedback, and improve it. Once you have a prototype and some initial customer feedback, that's when you should apply to Techstars.

And if you're an early-stage founder, I'm always available to guide you, and I can also refer you into the Techstars network.