Flock Safety has told the public that it operates more than 120,000 cameras across the country. But a new interactive map of Flock devices, published by a security researcher last week, puts that number at more than 170,000.
Now, Flock is demanding that the map be taken down.
The map is the work of cybersecurity researcher Joshua Michael, who in November 2025 discovered that Flock’s website was leaking the coordinates of all its devices.
But Michael’s map doesn’t just reveal the location of Flock’s cameras. It also pinpoints more than 130,000 accompanying devices. This can include the company’s 27,341 acoustic detection devices, used for detecting gunfire and other “critical sounds” like vehicle crashes and fireworks.
The map is different from those seen on websites such as DeFlock.org, which is based on submissions from members of the public who spot Flock cameras in the wild.
In remarks to The Intercept, Michael said he immediately alerted Flock to his findings last November. When he received no response, Michael says, he followed up with two more emails.
Flock answered after the third email and said it was looking into the matter. But the issue was only fixed after Michael published a blog post in January detailing his findings.
Interestingly, Flock published a blog post of its own in January, in which it stated that “Flock has never been hacked, and there has not been a leak of Flock information.”
Yet on Thursday — just one day after his map was published, and one day after his findings were mentioned during the Senate Subcommittee on Crime and Counterterrorism hearing on Flock — Michael received a demand that he take his map offline.
Michael said he was contacted by a company called Doppel, acting on Flock’s behalf, that accused him of violating the “FLOCK SAFETY” trademark. But Michael, whose website includes a disclaimer stating that it is “not affiliated with or endorsed by Flock,” has no plans to comply.
What I’m reading
- OpenAI’s Systems Meddled With U.S. Government Sites After Going Rogue: The New York Times breaks down how OpenAI’s artificial intelligence went rogue this summer and began poking around several government websites, including those for the Education Department, the Commerce Department and the Securities and Exchange Commission.
- Former Fort Hood soldier sentenced to nearly six years for hacking, extortion plot: The military-focused outlet Task & Purpose reports on a former soldier who was sentenced last week to 70 months in prison for working with a cybercrime group that conspired to hack and defraud at least 10 organizations, including AT&T.
- See how data centers are changing the American landscape: The Washington Post uses satellite imagery to highlight how farmland, deserts, forests and other rural areas of the country have been taken over by data centers. The Post says that around 60% of hyperscale data centers are built on previously undeveloped land.
What do you want to know?
That’s all for this week. If you have a question you’d like me to answer about privacy, send me an email at mthalen@san.com, reach out on Signal at @mikaelthalen.12, or share it below. I’ll keep all your personal details private.
More Stories from Mikael
- FBI’s secretive hacking unit exposed after breach by cybercrime group
- Why are your X posts disappearing? Platform says it is increasing transparency
- Hackers strike FBI, steal personal data that could compromise agents
To read past editions of Confidential Informant, click here.