Mass Surveillance, Police Misuse, and Who Controls Your Flock Cameras with Flock CEO, Garret Langley

Welcome back to the all-in interview, where we find the people building the future — sometimes a politician, sometimes a CEO, sometimes an investor, but always an important discussion. One of the most important discussions of the moment is balancing safety and privacy, and one company sits at the center of it. We're lucky to have Garrett Langley from Flock here. Garrett, we spoke on This Week in Startups, my other podcast that I've been doing for 16 years — that was six or seven years ago.

Six or seven years, yeah.

And I've been following your company.

Full disclosure: I tried to invest at one point. I'm not a shareholder, but I've always been intrigued by what you're doing on the margins, and sometimes concerned. Flock has been very successful with safety cameras—license plate readers specifically—and because of that success it has been adopted in many communities and has had a real impact on reducing crime. But there's now a group of people with privacy concerns, and we're going to go right at the controversies and this grand debate: how much privacy am I willing to give up to get safety? When I see that kind of energy, as an investor or as a podcast host, I know there's something at stake.

Tell us a little bit about Flock Safety, how many communities you're in now, and the drone component—people know the license plate readers, maybe not the drones. Let's build a mental model for the audience.

Thanks for having me, Jason. I started the company nine years ago with a pretty simple mission: to live in a safer neighborhood. That was our initial pitch. For our first two or three years in business, it was just neighborhoods—we sold a neighborhood a license plate reading camera, which is pretty basic technology. A car drives by, we take a still image of the vehicle, and we read the license plate. The system has gotten more sophisticated with time—we can now detect things like a roof rack, a bumper sticker, or a dent or damage—but that's at its core what the technology does.

Looking at last year, we've had a tremendous impact on the country: north of a million crimes solved using Flock, and north of 10,000 missing people found—everything from grandparents with dementia to children. Just last week in Atlanta, there was a stolen car that happened to have a 2-year-old in the back seat, and thankfully, because of very hard work by the Atlanta Police Department and our cameras, that little kid was safely returned within the hour. I don't know if that would have happened without Flock. That's one story, but it's my own city, last week. As a data point, in July there were about 1,125 missing people and over 20,000 stolen cars—and those are the cars we use to drive our kids to school and drive to work.

We're in just over 6,000 cities across the country. The consequence of being that consequential is that people have a lot of questions and a lot of debates, and I think it's worthwhile to talk about what kind of society we want to live in. I want to live in one where we're safer, and I want to keep my privacy while I do it—but finding that balance is not easy.

How Flock Cameras Get Installed

I remember when we had our first discussion on This Week in Startups, so many years ago, I brought up the question of abuse in the system—because once in a while you get a dirty cop. If you watch any procedural TV, Law and Order, the Sopranos, Tony Soprano has a dirty cop who's in the hole for gambling debt and runs license plates for them. It's a trope on the news, and it's a trope because it's common—not on a percentage basis, but we all know that any system that records data could be abused. We saw it at Facebook, where engineers famously stalked their exes and read their unencrypted messages, and the industry had to make a lot of changes.

At the time, you were telling me that the settings—and this is where we're going to get really granular today—are set by the local community, and these cameras are installed by a local community. The communities I've lived in have had them. So let's answer two questions: first, how does a community wind up having Flock cameras installed, and what's that decision-making process typically like? Then we'll get into very granularly what you're recording, for how long, who has access, the audit logs, and what's not recorded.

That's a great question. One of the reasons we struggled so much to raise money as a business is that it was traditionally very hard for Flock to get contracts done with local government. The last company to go public that serviced local government was over 20 years ago. It's a really hard space because city councils have the final decision.

Every two weeks a city council meets, and the city manager's direct reports—a police chief, a fire chief, public works—present ideas and concepts they'd like approval for. Flock is on that list. Last year we attended something close to 10,000 city council meetings as a business. I have a huge team whose entire job is going to city council meetings, explaining what Flock does and what it doesn't do. That's just the cost of doing business with local government.

I was actually talking a couple of weeks ago to our first-ever police chief, down in Jersey Village, Texas, and that council vote from six years ago is still public record.

City council meetings are famously recorded.

It's tough—it's a wacky environment.

It can be particularly wacky, but we go anyway, because we want people to know what they're buying. What has been true for the last nine years is that we've always said this data is a liability, not an asset. You should have a set retention—

The retention period determines how effective the product is. If you delete the data in a day or two, it's less effective—still better than nothing. Hold it for 30 days and it's pretty effective; hold it for a year and it's more effective. We've always believed city councils should have the final say on how long to store it and who to share it with. For the longest time, 30 days was our default, but we've recently shifted that down to 7 days. We think that's a reasonable trade-off, though we still think cities—or states—should decide. Someone who is democratically elected should have the final decision on how this technology works in their community. But as a technology provider, we have to set strong defaults—that's what we say in the industry.

Out of a hundred crimes of any nature, how many are solved with seven days of data, 30 days of data, or require more than 30 days? Do you have that?

We have a good sense. I'll give you an aggregate data point across the whole country: we think 90% of crimes will get solved within seven days of data retention.

Beautiful.

It is, Jason, but I probably talked to two or three hundred police chiefs this week, and they remind me—and I tell them they should tell their city council this too. The 10% we're talking about is the 10% you read about in the news.

It's the homicide where there was never a 911 call because someone was killed, and it takes five days to discover the body through a welfare check, and then it's a complicated case. And I'm sorry to go to this topic—it's a rape victim who felt guilt, who felt shame, and didn't call 911 until a week later because she wasn't ready. That's real life. I deal with this every day, so I'm probably less sensitive to it, but these victims deserve privacy and accountability for the people who attack them.

So to even identify the crime—of that remaining 10%, how many would be resolved with 30 days of data? We're just ballparking here.

We think the majority. And we're not going to tell cities how to run themselves. To be clear, there are states like New Jersey where state law mandates 5 years, so every customer holds it for 5 years. In California there's a maximum retention, and most states have now adopted regulation.

They set a maximum limit, but for me, more important than my default is making sure city councilors and mayors are educated that they have control. They can make this choice, and they should actually go spend time with their police chief to figure out what he or she thinks is best for that community. Some communities are going to push for two or three days, and that's so much better than nothing, because the Amber Alert and the Silver Alert—the things we all lose sleep over—will still get solved too.

I literally lose sleep over those. I got one at 2 in the morning the other day. I wake up, I read it, and I don't know what I'm going to do with this data at 2 a.m.—I'm in bed. But okay, this seems completely fair in terms of how you're expressing it.

To be clear, when I had you on Twist back in the day, I asked, don't you need to know who's in the driver's seat, and wouldn't you want a picture of the windshield? I found myself advocating in my community that I would be okay with it, just as Jay Cal the dad and somebody who sometimes on the margins has some privacy-security concerns—though not acute ones like, tragically, Sam Altman and what's happened at his house twice.

I have my own security cameras. I use Unity—it's incredible.

Great product.

And I have a license plate reader on mine, and facial recognition on my semi-private property. Of course, some of it does look out on public roads because I have access roads, so I'm technically able to record the license plate of everybody who drives by the front of my house. I think that's legal to do here in Austin.

Yes, it is legal.

"So, are there communities that are using other systems that are not Flock and recording every single person in and out? Because when I go in and out of private communities once in a while, I see there's four or five cameras. I know I'm on camera. So, aren't people already doing full video in their communities?"

"Yeah. Jason, you're spot on. We don't do facial recognition. We don't capture video. We don't look inside the car. We don't even allow in our system to search for people for our license plate cameras. And that's in direct conflict with some of our biggest competitors, who do all of the things I just mentioned.

"And unfortunately, or fortunately, we've been very successful. And as the industry leader, the spotlight's on us. And that's okay, because my hope is that the changes we're making, and the changes we'll continue to make, will force the rest of the industry to follow suit. And I hope counselors ask questions of the systems they have if it's not Flock. And maybe they'll switch to Flock if they want a privacy-first solution."

The company has faced a communication challenge: people have been ripping down the cameras, and not just in the United States—in the UK too. People care about privacy, and there's a group entitled to that opinion, which the CEO says he shares "on the margins." But when he had children, his view changed dramatically, because security became his number one priority. He's willing to give up his license plate data: if someone knows his exact whereabouts for 30 days from his plates, he's okay with that, because he has children. He would, however, be a little concerned about facial recognition, if he's being totally honest.

The interviewer pushed back on the company's reaction to the camera removals. The company had called the people taking cameras down "terrorists," but the interviewer argued they're obviously not terrorists—"activists" would be a better descriptor—and that the company hasn't communicated its position well. He asked where people are raising valid concerns, and how the company has evolved its communication. He noted that the people ripping down cameras seem to think they're living in a surveillance state with permanent records and facial recognition, and don't understand the nuances of the product. He asked the CEO to talk about where the company got it right, where it got it wrong, and how its communication has evolved.

The CEO admitted he's far from perfect and has made his share of mistakes over the last year—and over the last decade, a list he once tried to write down got so long he decided to focus on the future instead. On the vandalism itself, he finds it sad because these individuals get amped up on TikTok and Instagram to take action, and may not realize they're committing a felony. He mentioned an individual recently arrested in Ohio or Illinois who is facing 15 to 20 years in jail. He feels horrible about it—what if that person didn't realize the consequences? Meanwhile, the person on TikTok or Instagram pushing this isn't the one cutting the cameras down; they're doing it for likes and views, which he dislikes. Because when a camera is cut down and there's an Amber Alert, it won't work.

He still doesn't have much space in his heart for people who choose to commit a felony rather than go to city council and fight for what they believe in—they could even join city council. The interviewer noted that a number of cities have decided to take Flock down, apparently due to backlash, and asked whether the CEO is okay with that if it's the community's choice. He said yes, though it's disheartening when a city chooses to be less safe, especially when there are so many other options to reduce the product's efficacy besides removing it. To him, people who look only at privacy or only at safety are both on the wrong side of the equation: the point of the democratic process is finding common ground on complicated problems, and safety ranks high among what a community must provide.

He believes one of the most fundamental things government owes us is an equal level of safety. Today, depending on your skin color, wealth, race, and sex, how safe you feel varies: women feel less safe than men—that's a fact—and minorities feel less safe than the two of them. If you live in a less affluent city or a less affluent part of a city, you feel less safe. He believes government has an obligation to make everyone safe, and that's what Flock is trying to do.

Vandalism is an easy case to justify. But there's a nuance here: someone might say, "I didn't opt into this. I don't want my license plate in there." Is there an ability to opt out?

What is the maximum retention period for the data that's saved? And then, really, this is a very nuanced point: what's the audit trail? If there's a dirty cop who's into Tony Soprano for a hundred dimes betting on NFL games and wants to hand the mob a bunch of license plate data from Flock, what's to stop the dirty cop? And what's the largest search window people can use today on Flock?

Let's hit on the abuse issue, because I've been called naive about it—I'd say maybe optimistic. I grew up in Atlanta and always trusted the Atlanta Police Department. Over the last year, I've spent a lot of time talking to people who don't like Flock, and I finally got to the bottom of why. I kept asking, why do you hate this company so much? We sell a camera. This woman told me, "Garrett, I don't trust the police at all." That was the root issue. When I asked why, she said, "They lie, they cheat, they steal. I don't trust any of them." I thought, okay, what if you're right and I'm wrong?

She's right. There have been many cases of police departments subverting even wearable cameras—they turn them off. We have ICE agents wearing masks with no numbers. So the public is rightfully concerned, and we're bundled in as their partner. That's where they have a valid concern.

I think if you make it a more macro topic: anyone in a position of power—CEOs, elected officials, police—there's a higher likelihood of abuse. When we met six or seven years ago, we had the audit log, and my assumption was that this was good enough, that it's what most police systems do. What I didn't realize was that he was meeting with a large agency in Southern California with something like a thousand officers a day using Flock, each doing five or six searches a day—and once a month, one person is supposed to review that log. They'll spot-check 10 or 20 searches, which is the best they can do, and move on.

So four months ago we built a new tool called audit assistance. We solved a million crimes last year, so we have a good sense of what normal usage looks like, and with just under a dozen abuses detected over the last eight years, we also know what abuse looks like.

Here's the easiest example of abuse. Say you're an officer trying to stalk me. You'd search for my car—plate ABC124—figure out where it is, then search again the next day, and again the day after. That's very abnormal. If you were truly trying to find me because I committed a crime, you would have put my tag on the hot list so anyone else in the department would be notified too—you might be asleep or off shift, and everyone has a right to try to find me. If you're stalking me, you don't want anyone else to know what's happening. That's the easiest one we can flag.

We built this tool four months ago and put it out there not knowing how well it would work or how much signal versus noise there'd be. You've seen the headlines over the last few weeks—we've caught a lot of bad cops. More than I ever would have hoped. In my home state of Georgia, where some of our customers have been with us for years, I have to applaud those chiefs because they're not hiding behind it. They're publicly saying, "I fired nine officers today—the most severe offense in Georgia for abusing the system. They're out of my police department. I don't want them."

I think that's good. We shouldn't be criticizing those chiefs; we should be saying thank you, and who's next—who else wants a clean police department? So we're now requiring audit assistance as a mandatory feature, not even just a default. It has to be turned on. We have to find these bad cops and get rid of them—they shouldn't have a badge.

This is interesting because if we assume, just to pick a number, that 1% of cops do bad things—some people might think it's all cops, some might think it's 10%—

I'll just say 1%. It's a very small percentage, but in a city with 30 or 40,000 police officers like New York City has — and I think Los Angeles has 10 or 15,000 — 1% suddenly becomes 100 or 300. That's a lot of bad cops. And you've set, in a way, a honeypot for those bad cops. They probably don't even know this is being tracked.

My question is: when I make a request as a rank-and-file cop, why should rank-and-file cops be able to do this? Shouldn't there be a higher-level person who handles the request? Maybe a supervisor says, "I'm looking into this case," explains why, and then a senior person like the sergeant on duty has to approve the search. Then a third person reviews the sergeant's approvals and adds a note. You could build that into the system — a double-key system where two people have to say yes. Now, as a dirty cop, you'd have to find a dirty partner.

What you're describing is definitely on the whiteboard for how we continue to curb abuse — I can show you screenshots of what it would look like. The challenge we face — and you've dealt with this in your software companies — is that we have some customers with thousands of employees, and some like the police chief in Texas I was just on the phone with, who has 20 employees. The burden of what we just described would be pretty big for him. With just 20 officers, he's also not using the product every day — he doesn't need to, there are only about 2,000 people in the town, so the audit log is very slim. He doesn't need any of the things we're talking about.

So defaults do matter, and this might get to a level of nuance where big cities have one set of rules, medium cities another, and small towns different rules — no different from how it's harder to become a police officer in New York City, with a longer academy, than in a small town, because the expectations are different. So yes, you're spot on: we're not done. Abuse is an industry-wide problem, not just a Flock thing.

Look at California: there were 7,000 reported abuses of their state database last year. Seven thousand. Which means you should expect some percentage of that in your system — the Washington Post did a story on this. The company is evolving and reacting to that. As a technology provider, you're now responsible for the worst behaviors of your customers. The lesson is that even if you want to say "we just sell a tool," the reality of running this company is that people give up some level of privacy to get that security, so you actually have to own it. Your previous approach was "it's a tool, do with it what you will" — you made that choice.

Which isn't exactly unreasonable. People sell laptops, cameras, guns, and tasers to local police departments, and we don't tell the taser company, "You're responsible when somebody gets tased inappropriately." But here we are — this is qualitatively different, because Americans really value privacy.

Sales tax rules are changing: starting January 1st, California will tax certain software for the first time. Numeral is the sales tax solution for fast-growing businesses, handling registrations, filings, tax rates, and compliance across the US and over 70 countries, with Numeral's guarantee. Learn more at numeral.com.

To your point, my view of our responsibility has evolved. I would prefer if every elected official understood how all this technology worked and made decisions that represented their community values. But I don't think that's a reasonable expectation — it would mean they also need to know everything about data centers, everything about water works. They rely on their police chief to tell them what's best, and that chief relies on me to educate him on what I think is best.

And so I do think the responsibility falls on us to give those strong defaults and then allow them to pivot against it, still respecting the democratic process but focusing on the defaults.

Yeah. And as the lead candidate, you're going to take the arrows. But there are other people. My friend Jamie Simonoff, who's an incredible entrepreneur and a dear friend of mine for many decades — he has Ring, famously. During the Super Bowl, he did this thing like, "Hey, with Ring cameras..." I guess I don't know if they're considered your number one competitor, because they sell to individuals and then allow the data to be used by police if people opt into that, which is an individual's right.

They were like, "Hey, we can find a lost dog." And people lost their minds over that, because they said, "Hey, if you can find a lost dog, then you can have somebody stalking" — whatever people's valid concerns are. And my understanding is they had to turn that feature off after releasing it.

Yeah. I've known Jamie for a long time as well. I don't think they're a competitor; I'd say they're a friendly. We have the same mission, which is to build safer communities in America. And I think he and I have both learned a lot over the last year about maybe underestimating people's concerns with how technology could impact them, and needing to go slower in some cases.

We haven't gotten into the AI topic yet, but it's an area where we're doing it smarter now. There's a version of Flock where we would have just ripped and launched AI feature after AI feature and said, "This is awesome." Instead, we're going intentionally slow.

What's the opportunity and the concern with AI? Give us the practical picture. If we were in a police state — if it were China, where the CCP runs a literal police state — that's very triggering for Americans. They see that on one end, then see Flock here and Ring there, and ask, "What's the distance between these two?" AI is the obvious answer, because there they're tracking people with facial recognition and watching their behavior: "This person littered. This person raised their voice." It's pretty gnarly.

So tell us about the AI features — what the detective wants most, since the detective wants to close cases — and then what you look at and say, "Gosh, that would close the case, but maybe it's too much of a privacy concern."

Let's talk about a simple one that's in debate right now: 911 systems. Today, when you call Delta, maybe you get an AI agent, maybe a human — but you're booking a flight, so the risk of hallucination is pretty low. Maybe they book you on the 7 p.m. instead of the 4 p.m., and you're like, "What just happened?" But it's fine. Now imagine someone broke into your home, you call 911, and the system has been switched over to an AI bot that starts talking to you about what you had for lunch. No good. The risk is so high, we have no idea how these things work, and there's no good way to control them. I think that's pretty well established. Yet cities, for budgetary and staffing reasons, are leaning in aggressively, and there have been some bad cases where the AI got it really, really wrong.

There's a smarter way: have AI pick up only when the phone lines are fully busy. That would be smarter — but that's not what's happening. The reckless way is to just flip a switch.

Do you handle 911 systems?

We don't do 911 calls, but I'm watching this unfold and thinking, this is a huge risk.

I suppose you could use AI to look at patterns and flag abnormal ones — a car slowly driving up to a house, driving away, then doing the same at the next house, six times. Is this somebody delivering a lot of burritos, or somebody casing the joint?

That's what the detective would want — precognition from Minority Report, predicting a crime before it happens.

That I don't want to do. We've been really consistent that we never want to be a company that defines suspicion. That's really dangerous. It's an interesting happy-hour topic of what that world would look like, but for us, it's really valuable that there's a human in the loop when it comes to public safety. We're talking about people's lives — this is really important stuff. We always want a human in the loop.

There are some products we could build, but the difference comes down to knowing the cost of getting it wrong. You'd have to deploy this with third-party attestation that it's working as expected. It shouldn't be up to me, and it shouldn't be up to my customers. Someone who doesn't actually want this to work should come in and verify: "Guess what — it does work. It's making officers more effective and more objective, less subjective." That should be our end goal: AI used in a way that increases the number of crimes solved and decreases the likelihood of getting the wrong person. That's an easy ideal state to describe — I just don't think we should rush there. We should be really deliberate about how we get there.

You've lost, I think, dozens of cities out of the 6,000. Some cities are now saying this isn't worth the trouble, or too many people protested.

Take me through what's happened over the last year with those cities — I think it's 60 or 70 out of 6,000, if I have my numbers right, about 1% that have said, "Okay, this isn't for us."

It's tough, because I want everyone to be safe — that's why I started the company — and I know what these cities have had and the results we've delivered. Unfortunately, most of the time they're making a decision based on misinformation. I remember vividly one of the first customers that chose to walk away did so because of facial recognition. They went online and googled "ALPR facial recognition," landed on a competitor's website, and said, "We're getting rid of this Flock thing." They didn't realize they were looking at the wrong company's website.

And that's tough, because that's on me — it's on me to better communicate to these decision makers: here's what we do, here's what we don't do. That's been the bulk of it. The other thing that's been hard to see, Jason — there was one city in Washington, and it's just so sad. They turned the cameras off, and the next weekend a little boy gets shot and the car drives off. It's going to be a cold case. Luckily, the city next door still had Flock cameras, found the car, and made the arrest. The kid didn't live, which is horrible — at least his family knows justice was served.

So we're making really consequential decisions on inaccurate information, and that's tough. We have to do a better job as a company communicating to our customers, and to their customers — which would be the community — about what this technology does and what it doesn't do. When I speak to, let's say, 99 people out of 100 and explain the facts, they say, "Why is this even a debate? We have control as a city. We can set limits, we can set guardrails. I want to be safer than I feel today." I've never met someone who says, "I want to be less safe."

I predict that as you explain this, change the defaults, get the audits in place, and get the dirty cops taken out of this, some number of those cities and city councils will revisit this when things cool down a bit. It happened where I live now, in Austin, Texas. Austin canceled Flock — or didn't renew, whatever — in 2025, according to my notes. People have guns here; I have a gun, I carry a gun, everybody's got guns. We had this crazy shooting involving a bunch of kids, similar to what you're describing. They were committing these crimes in Austin proper, then drove out of Austin, and the Manor Police Department did have Flock. They picked the car up, and 28 hours later the three teens were arrested — juveniles, 15, 16, and 17, according to what I'm seeing. That was the wakeup call here; it brought the debate back to the front, and now everybody's debating it again in Austin.

The ACLU wants 48 hours, I think, as their proposal.

That's reasonable. That's the privacy organizations, the civil liberties groups — we get it. Most cops probably want 30 days. Communities are going to set it at seven or 14 days. That's it.

You have competitors. What's their approach here, and how do you deal with that as an entrepreneur? Are they just saying, "We'll sell you the technology and keep our heads down" — or do they market themselves as "Flock without limits"?

To your earlier point: we've already had close to 20 cities turn their cameras back on.

The speaker notes that close to 20 cities have already turned their cameras back on, because they realize they are now less safe and have an obligation to their constituents to be safer.

It is tough as an entrepreneur, he says, because you want to build as big a business as possible, but you also want to do it in a way that, when it's all said and done, your kids say, "I'm so proud to be that guy's son or daughter." So you have to make decisions you'll be comfortable with ten years later. He had a lot of concerns about requiring accountability from all customers, knowing he might upset many of them. But he told himself: if a customer doesn't want to be held accountable, he doesn't want them as a customer. They can go to a competitor who doesn't have these tools or requirements, and that's okay — it's just business, though a business serving a very important mission. He wants to sleep at night, and he does. His competitors, by contrast, run more recklessly without these tools or scrutiny, and he believes it will catch up to them eventually.

The interviewer agrees: you'll lose some customers who want an untethered system without proper constraints, but you don't want them anyway because it will blow back on you — as happened famously in San Francisco with ALPRs, automated license plate readers.

What the cameras capture

Asked whether the system also records dents, color, make and model of cars, he confirms it does. The technology is effective because on a 911 call, most people don't instinctively remember a license plate — they remember what was unique about the car, like damage on the right side. He recalls a homicide case near Atlanta where all the witness could remember was that the back right tire was missing a hubcap. Detectives then asked to see all cars in a 15-minute window without a back right hubcap — that's the beginning of an investigation, not the end, just a way to get started. That was all they had, but a very good detective worked hard and found the person.

The "Flock effect" in San Francisco

He saw a tweet recently about "the Flock effect" — crime down, stolen vehicles down in San Francisco. But Flock is only one small part of a much bigger transformation in that city, including stronger elected officials, stronger enforcement of the law, a strong police chief, a strong DA, and a community that demands to be safe. That's what should happen if you live in San Francisco.

If you live in San Francisco, you deserve to be safe. It's one of the best cities in the country, and it has a lot of problems. People there are very libertarian, privacy-based, live-and-let-live. Having lived there, I think that openness is part of the problem — it literally comes from the Summer of Love and the fact that it was a mining town. A lot of the culture that emerged is "let your freak flag fly, do what you want to do" — even to the point of asking, if people are stealing, why are they stealing? What happened in their childhood? People there have had enough.

I think the correlation is this: do you have the ability and the entitlement to say, "I don't need to worry about this because I have security at my house, a security guard, a personal driver, personal security"? A lot of rich people now have some combination of these things at a minimum — I have cameras, I have people, all of this. You have to be extremely aware of your own privilege that you don't have to deal with crime.

My brother was a cop during the stop, ask, and frisk era in New York. The statistics during stop, frisk, and ask — which was a violation of the Fourth Amendment — you were stopping people just for hanging out, so it definitely went a step too far. But New York in the 70s and 80s was dangerous as hell. There were a lot of guns on the street; my brother took a couple off the street. Cops were in a really tough situation, and the people in the poorest communities were the most in favor of stop, ask, and frisk. They always leave out the "ask" — no, they would stop and say, "Hey, what are you up to? You're hanging out on the corner, what's going on?" They just tried to get a read on people, which, by the way, is what happens if you go to Israel or other countries: they ask you 20 questions — why are you here, where are you staying — whether you're entering Israel or Tokyo. They might ask a couple of personal questions that would be offensive to Americans.

But the correlation was real: people who experience crime are in favor of it. People who live in Park Heights don't need to worry about this — if something goes down there, there are 20 cops and 50 private security guards rallying around one homeless person, one vagrant, one person in a car that doesn't look like it belongs. They're going to get swarmed in Park Heights.

You hit the nail on the head, and it's not spoken about as much: it's worth looking at who is so against what we do and who is so for it. It's typically people who don't have any other way to protect themselves — who need this, who want this. For those who have a gate, a guard, cameras, a security system, a doorbell camera, all these products and services, they ask, "Why does anyone need to feel any safer?" The answer is: because most people don't have any of this stuff, and they don't feel safe when they go to bed. It's quite telling.

I'm rooting for you to get everything right and to continue making communities safer. I'm also rooting for communities to make their own decision.

I think that's the key part people need to embrace. If you've chosen to live in a gated community, you have made that choice for security in all likelihood.

And if you ask poor people, or people who don't get to live in a gated community—people with a family—"Would you like to live in a gated community or not?" I'd say a hundred out of a hundred, maybe 95, would say yes. If you offered the average Joe or Jane a community gated with three gates and a security guard, they'd say "heck yes—but I can't afford that. How much is that going to cost?"

So people with children and people with homes just have a different view of this than when you're younger and can afford to say, "I don't have kids, and I care about my privacy, not my safety." And I understand that. I don't want the government to have a national Flock database. I don't want the government doing facial recognition—I find that a bridge too far.

But license plate retention for two weeks or 30 days? I kind of feel okay with it. I'm sorry if people are disappointed in me for that, but as a parent and as somebody in a community, I'm okay with my community having license plate readers for up to 30 days. That's the decision I came to.

To your point, Jason, I had an interesting conversation with an elected official in Illinois who was one of our largest antagonists—very anti-Flock. Then something really scary happened: she was the victim of political violence. Someone shot at her house, and the police used a Flock camera to find that person. To her credit, she changed, and she's now one of our largest advocates, because she realized you don't get to pick when you're the victim of violence. Sometimes it's random, sometimes it's intentional, but everyone wants justice when something bad happens, and everyone wants to know they're going to be safe.

So I think you're right, and my hope is that these kinds of conversations lead to fruitful debates about how we can have both—without necessarily letting a vocal minority dictate us moving toward a less safe society. That would be pretty sad.

Asked how far through the PR disaster Flock is and whether the company can recover, the CEO said the biggest damage has been to internal morale. Employees go on X or Reddit and ask why people are so angry at a company that has been doing the same thing for nine years—and angry about things Flock doesn't even do. "We just want people to be safe," and having your work challenged so viscerally is hard.

On the positive side, the backlash has brought the company closer to its customers, whose job is keeping their communities safe and who see this technology as one of the most—if not the most—critical parts of doing that job. One police chief called it the best thing since DNA for fighting crime, and he agrees with that assessment.

The business is in good shape: Flock raised more money at the beginning of the year and is still growing, though it could grow faster. The company used to have zero churn; now it's around 1%. He'd like to get back to zero and intends to fight as hard as he can to get the right message out—communicating better and listening more. If they do that, they'll be fine, because everyone wants to be safe. That hasn't changed and probably never will, just like everyone wants food and a roof—core needs. As Maslow's hierarchy puts it, safety is up there.

The interviewer pushed that the company will have to lead and stay ahead of these issues, citing Zuckerberg and Facebook as a cautionary example: they've always played defense and catch-up on Instagram's impact on young teens—eating disorders, social media addiction—while X and Twitter dealt with political subversion. Every social network had a serious downside that's now been quantified, and it would have been far better to get ahead of it. Zuckerberg's recent manifesto earned some credit: despite obvious selfish business needs and wanting to be part of the AI revolution, he seemed to be getting ahead of the problem and showing empathy—acknowledging people could lose their jobs to AI, arguing that using the tools can prevent that, and that open source, free agents, and free AI that others pay thousands of dollars a month for might level the playing field. That's what CEOs have to do now in a media-saturated environment where a story can go viral instantly: anticipate it, own it, and be frank about it.

Asked whether he reached out to Tucker Carlson, who "demolished" Flock last week, to go on his podcast, he said no—he doesn't know if that's in his deck of cards. He does like t-shirts, though he put on a polo with a collar for this interview. More broadly, the company appeared on the Today Show and Good Morning America that morning; they want to go where the community listens and have meaningful conversations. They're not done—there's a long road ahead—and they didn't even get to drones yet, which he notes is probably Flock's fastest-growing business unit right now.

Drones, Transparency, and Accountability

Our LPR business is just under 50% of our forward-looking revenue, so it's a big part of the business, but we also have drones, live video cameras, and some really cool software. The drone story is one of my favorites: these drones live in an HVAC-style control box—a dock—on top of a roof, typically at a police precinct or fire department. When a 911 call comes in, instead of sending an officer first, you can send a drone.

There's a case not far from you in the Dallas–Fort Worth area where a woman called 911 late at night from a gas station and said, "There's a man walking back and forth with a gun. I don't feel safe." It's not illegal to carry a weapon in Texas, but she was scared.

Normally what would happen is one cop would hear that, flip on the blue lights, race down there with his heart racing and his gun ready, and someone's about to die.

Instead, they launch the drone. It flies at 400 feet at 60 miles an hour, so it arrives in under a minute as the crow flies, with a 40x optical zoom. It zooms all the way in—and it's not a gun. It's a lighter, one of those toy-like lighter things. So they call the woman back and say, "Ma'am, it's just a lighter. You're okay."

I share that story because so much of my businesses have revolved around solving crime, but it's more than crime—it's about safety. How do we help the community feel safe? Here we mitigated what would have been a really bad situation for everyone. We have thousands of stories like that where communities are safer with our drones and our cameras. And to your earlier point, we just have to do a better job communicating how it all works—and teach our customers to do that too, because we're only one company with thousands of customers, and we need their help telling the story.

The interviewer suggested that the audit-trail tool has great potential: a third-party auditor, the way people use Ernst & Young to audit books, could verify that a drone launch was legitimate—that it wasn't a dirty cop sending a drone to stalk an ex-spouse or invade someone's privacy. Every drone flight could be reviewed in real time, because once drones start flying, Tucker Carlson or the ACLU will look at it and say you're peeking in my window. There have to be massive safeguards: is a drone deployed only if there's a 911 call?

This is the next big chapter for law enforcement: implementing drones better than they've done with historical products. Look at body cameras—there was massive outrage at first, and now people say, wait, this is actually a really good tool. Same with non-lethal weapons. There's still more work to do. On third-party auditors: you've got a city manager whose job is effectively to hold the police chief accountable, and they can go into the audit logs themselves.

They can see the audit logs and the flight patterns. Most of our customers have a transparency portal with the audit log, how many cameras they have, their drones, and their drone flights. There's a lag—we want a delay, for instance during an active criminal investigation—but what will be hard for our customers is getting comfortable with this ever-growing expectation of both transparency and accountability. We're going to push them hard, but the community is going to push them just as hard.

And some of these things are already public: 911 calls in many jurisdictions are public record.

They're public record. Same with radios, although some places don't treat those as public record—they encrypt the radios. Drone flights are one area where the question arises: should drone flights just be posted on the website—here are all the drone flights that occurred, along with the 911 calls connected to them? Does your software do that for them?

It would be possible. We've never had a customer ask for it. We have a concept of a spectator mode that would be available to city officials and command staff. But I think your question is the right one. As an industry, we're going to have to make these decisions about what the default is sooner rather than later. We have a default now, but there's very little state regulation and zero federal regulation outside of air safety—which is important, but nothing yet at the state or city level.

So you've got New Hampshire at three minutes—three minutes for license plate reading. That's New Hampshire, though. And then New York state is 21 days. That's the state of Washington.

Well, there you are, folks. If you want to really live off the grid in privacy: New Hampshire. Live free or die. Garrett, I appreciate your candidness, and good luck being in such a controversial space and getting ahead of these things. We can agree publicly to agree and disagree on the margins about these things, but it's up to the community to make these decisions—and I think that's the high-order bit here. What you're starting to get right is: this is your community's choice. You don't want it, turn it off. You want it for 48 hours, set it at 48 hours. We'll have some reasonable caps on this. I wish you great luck, and thank you to everybody for tuning in to the All-In interview program. Bye-bye.