How I decided to leave the FTC to run Signal

Almost a year ago, Meredith Whittaker announced she was heading to the Federal Trade Commission as a senior AI adviser. But the outspoken tech worker activist and AI researcher knew she wasn’t interested in government long term. Last month, Whittaker announced her new role as the president of encrypted messaging app Signal.

Signal has become an incredibly important app for activists, cybersecurity experts, and other folks committed to private, encrypted communications. Brian Acton, co-founder of both WhatsApp and the nonprofit behind Signal, is interim CEO; the organization is still on the hunt for a permanent CEO, Whittaker said. As president, Whittaker will be in charge of Signal’s policies, image, and strategy. Her priority is building a sustainable business model for Signal without compromising its ethos of privacy and security. Signal doesn’t have solid plans yet, though Whittaker is looking into a small donation model. “This could help create a template for models of building tech that aren’t based on the surveillance business model,” Whittaker told Protocol.

It’s always a hefty battle convincing users to start paying for something they’ve taken for granted as free, especially when competing against the likes of Meta and Apple.

“One thing that is important to make really clear as widely as possible is how expensive it is to create and maintain these systems,” Whittaker said.

Whittaker has had a varied and storied career, first coming into the public eye via her protests as a Google employee against the company for building AI technology for drones, among other things. She co-founded the AI Now Institute at New York University to study the social impacts of AI, leaving Google eventually to run it. In 2020, she joined Signal’s board of directors. She describes the move to step up as president as slow-moving and organic. In some ways, she’s been “accidentally training for the role” her whole career, she told Protocol in an interview about her decision.

Whittaker’s story, as told to Protocol, has been edited for clarity and brevity.

We could draw this back a decade at this point. I have known [co-founder and former Signal CEO] Moxie [Marlinspike] for almost a decade. I was using Signal before it was called Signal, when it was still RedPhone and TextSecure. I’ve been deeply invested in Signal and an unequivocal champion of what Signal does and the necessity of Signal existing and thriving.

My joining Signal as president was a conversation we had a handful of times over many years. For me, there’s an element of, “you know it when you know it.” And I just knew this was the right moment.

The decision was very well informed, probably more well informed than any other career decision I’ve made. I had such a privilege to have real insight into the organization, the issues specific to Signal, and the project of developing a truly private communications infrastructure. I had co-founded and led AI Now for five years at this point. I did a term as a senior adviser to Lina Khan at the FTC. I’d been at Google and had a long run and a lot of experience with the brass tacks of product development. What are the actual practices inside a tech company? What are the dependencies and issues that you run into? I was in academia for half a decade and did a lot of work looking particularly at the political economy of tech and the tech business model.

It seemed like the right time. We’re at a moment where the surveillance business model has calcified. We have a handful of big tech companies. They control the infrastructure, they have the capacity to continually gather data constantly. You have a service like Signal, which I believe is existentially important for a livable future. We need to be able to communicate privately. We need to be able to experiment with ideas. We need to be able to discuss a health problem without fear that our employer is going to surveil that discussion and cut us off of health insurance.

The stakes vary from the very mundane to the geopolitical. Nonetheless, you have to be able to communicate privately or you basically live in a world where the power relationships that exist are cemented and immovable.

We need to be able to communicate privately. We need to be able to experiment with ideas. We need to be able to discuss a health problem without fear that our employer is going to surveil that discussion and cut us off of health insurance.”

I saw this industry metastasize from the inside. I’ve been in the tech industry for 17 years. I started at Google in 2006. Yahoo was a bigger site by monthly active users than Google when I joined. Google still used Outlook internally as its email client because Gmail was in beta. It was a very different time. The iPhone didn’t exist, we didn’t have [Google] Maps. I had to print out MapQuest directions to get to my Google interview. I had a very close view of the surveillance business model’s ascent.

One of the key things that I am looking at is sustainability. How do we build robust, always available global communications infrastructure without participating in the surveillance business model? We’ve had a couple of decades now of “free” products that were not actually free. They were monetizing data behind the scenes. There was a sort of shifty trade-off, and now we have a world that is interpolated by mass surveillance at almost every turn.

Signal is a free product because we believe everyone should be able to have private communications. It shouldn’t be something that is a luxury only for the people who can afford it. That’s not how friendships work. That’s not how relationships work. We want to make sure anyone who wants to talk to anyone privately can pick up their device, install Signal, and do that.

However, just because we reject the surveillance business model doesn’t mean it’s any cheaper for us to develop high-availability, always-on software globally. It costs us tens of millions of dollars a year, and that’s a really lean budget compared to a lot of the alternatives. Signal is in total 40 people. Those tens of millions of dollars a year are going to hosting, registration, the basic costs of just keeping a service like Signal alive.

We’re really lucky at Signal because we have a long runway to experiment and really get the sustainability model right because we have the generous contribution from Brian. But we’re still looking at tens of millions of dollars a year and we need something that can meet that need.