Intertrust.

A small office on Shattuck Avenue, three blocks from the Berkeley BART station, is humming through another Tuesday. Inside, engineers are pushing certificates to a smart inverter sitting on a roof in Adelaide. A few desks over, somebody is signing a build of a forensic watermarking SDK that will end up inside a major streaming app you almost certainly use. None of this will make the news. That is, more or less, the point.

This is Intertrust. The company is 35 years old, owns a famously deep patent portfolio, employs around 160 people, and has the same CEO it had when George W. Bush was in his first term. It does not have a viral logo. It does not run Super Bowl ads. It sells a product called "trust" and somehow makes a living doing it.

Before "the internet" was a noun, they were worried about it.

In 1990, an entrepreneur named Victor Shear started a company called Electronic Publishing Resources. The thesis was straightforward and, at the time, mostly ignored: digital content was about to leak everywhere, and nobody had a way to track, license, or protect it once it did. Music had not yet been ripped. Napster did not exist. There was no streaming. The pitch sounded like science fiction with a tax problem.

By 1996, the company had renamed itself Intertrust Technologies and started filing patents at a furious clip. The patents covered something the industry would later call digital rights management - the cryptographic plumbing that lets a song, a film, or a database row travel through an open network and still know who owns it on the other end.

Translation: somebody had to think about copy protection before there was anything worth copying.

One bet, repeated five times.

Intertrust's history reads like a single conviction running through multiple internet eras. The bet was that decentralized systems would always need a centralized story about who is allowed to do what. The internet would grow. Devices would multiply. Content would scatter. Energy itself would become a packet. None of it would work without somebody quietly handling the keys.

The company is unusually patient about this. In 2003, Sony and Philips - along with Stephens Inc. - bought Intertrust outright, taking it private and bringing in Talal Shamoon, then a researcher at the company, as CEO. He has held the job ever since. Twenty-plus years in a chief executive seat is, in tech terms, geologic.

In 2004, Microsoft paid roughly $440 million to settle a long-running patent dispute with Intertrust. The settlement was, at the time, one of the largest in technology. The press coverage was brief. The company moved on.

What you actually buy from them.

Intertrust does not sell one thing. It sells one idea, broken into five product lines, addressed to three industries that do not normally talk to each other.

Five products. Same underlying chassis. Five very different sales calls.

Who is paying for this.

Intertrust does not publish a customer list in the way that a SaaS company would. It does not have to. The names that surface in case studies and partnership announcements are the kind that close doors on followups: Sony, Philips, LG, BT, E.ON, Origin Energy. Plus a long tail of utilities, broadcasters, and equipment makers who would prefer not to be quoted explaining how they handle device certificates.

Building trust for a connected world.

The phrase is, on its face, slightly ridiculous. "Building trust" sounds like a poster in a midwestern bank lobby. But strip out the corporate sheen and the line is honest. What Intertrust sells is, at root, the ability to know - cryptographically - that a thing on the other end of a wire is what it claims to be, allowed to do what it is doing, and accountable for doing it.

This is dull to read about and hard to do. It is also the reason streaming services do not collapse the moment a user starts a video, why a solar inverter on someone's roof can talk to a grid operator without becoming a vector of attack, and why an industrial sensor in a refinery can be retired without leaving a working set of keys behind.

Intertrust's mission, then, is to make these guarantees small, reliable, and boring. Boring, in this industry, is the highest possible compliment.

Boring is a feature. Discuss.

The grid is the next streaming service.

If you squint at the energy transition, it starts to look like the early days of streaming media. Millions of devices. Distributed authors. Real-time exchange. Money on the line. Regulators paying attention. A flood of new participants who do not have a security team. Intertrust's argument is that the grid will need exactly the kind of certificate infrastructure that, ten years ago, the media industry had to build under duress. The company is now selling that infrastructure in advance, to utilities that are not yet sure they need it.

Then there is AI. Trusted data exchange, model provenance, signed inference - all of it sits squarely inside Intertrust's worldview. None of it is glamorous. All of it is increasingly required.

Back on Shattuck Avenue, the Tuesday is ending. The Adelaide inverter has its certificate. The watermarking SDK has shipped. Somewhere, a regional utility is reading a Flexworks dashboard and deciding whether to dispatch flexibility from a virtual power plant. Somewhere else, a film studio is signing rights into Marketmaker. The office lights go off. Trust, the thing the company has been quietly selling for 35 years, persists in its absence.

That is the trick. Intertrust does not need to be loud. Its product, when it works, is the silence afterwards.