The Story of Hush: Building the Future of Privacy in the AI Era

Sometimes the most valuable insights come from seeing the same problem from different angles. In a recent episode of Category Visionaries, Mykolas Rambus shared how his journey through multiple startups – from the dot-com boom to the 2008 financial crisis – shaped Hush’s approach to reimagining privacy in an AI-powered world.

From Crisis to Opportunity

The story begins in the aftermath of the 2008 financial crisis. While working at Wealth X, Mykolas observed a crucial shift in the financial industry: “UBs, Citi, all the big names are talking about wealth management in the future of where that was going.” This led to building what he describes as “a Bloomberg for private bankers,” aggregating publicly available information about high-net-worth individuals.

But it was during this process that Mykolas discovered something surprising: “Most have no idea, right? Most presume that the donations they’ve given or the quotes that they’ve made, and certainly information about their wealth or how they’ve created wealth are not as easy to find.”

The Equifax Turning Point

The real genesis of Hush came during Mykolas’s time at Equifax, where he ran the consumer data division “before, during, and after they went through the big breach.” This experience, combined with his insights from Wealth X, highlighted a growing vulnerability in corporate security – the human element.

Building from Detroit

After leaving Equifax, Mykolas made a deliberate choice to build Hush in Detroit. “I very much had the itch for a few things. One, to build what was next. Two, to do that in my hometown of Detroit, where I’m based now. And three, solve a problem which we had personally dealt with at Wealth X and which I saw no one solving yet.”

This decision wasn’t just about location – it was about tapping into what Mykolas calls “a special kind of grit that comes from not only the midwest work ethic.” This foundation has proven crucial as Hush tackles increasingly sophisticated threats.

The AI Inflection Point

The timing of Hush’s approach has proven prescient. As Mykolas notes, “There are a lot more impersonations than there ever have been. The toolset is increasingly inexpensive. We’re talking $20 worth of software and a laptop in the world.”

The rise of AI has only accelerated this trend: “This is going to be one of the biggest issues around online identity and safety in the next 15 to 20 years.”

Democratizing Privacy Protection

What sets Hush apart is its approach to democratizing privacy protection. As Mykolas explains, traditionally, this level of privacy protection was reserved for the ultra-wealthy: “In the past, they would hire folks who used to work at three letter agencies, who were on retainer… That was tens of thousands of dollars per year, if not more.”

Hush has made this level of protection accessible to everyday employees: “We’ve definitely democratized that process, making it accessible for any given employee at a company to have that level of protection.”

The Future of Privacy

Looking ahead, Mykolas envisions a fundamental shift in how society approaches privacy: “I would love, and our company would love to see the nature of privacy change, not only in the US, but also globally.”

The goal is ambitious but clear: “Our hope is that at least in the next 3-5 years, 20% of the american workforce is protected by something like hush.” This isn’t just about corporate security – it’s about reshaping expectations around privacy in an increasingly connected world.

For Mykolas, this mission extends beyond business metrics to fundamental questions about privacy in the digital age: “It’s a reasonable expectation for any American to know that when someone drives down the street and types in their address, types in a address, they won’t be able to figure out who lives there, all about them, in two minutes flat.”

In an era where AI makes impersonation increasingly accessible, Hush’s story suggests that the future of security might not lie in building better walls, but in removing the target altogether.