Flock’s Gunshot Detection Microphones Will Start Listening For Human Voices
from the surveillance-everywhere dept
Flock Safety, the police technology company most notable for their extensive network of automated license plate readers spread throughout the United States, is rolling out a new and troubling product that may create headaches for the cities that adopt it: detection of “human distress” via audio. As part of their suite of technologies, Flock has been pushing Raven, their version of acoustic gunshot detection. These devices capture sounds in public places and use machine learning to try to identify gunshots and then alert police—but EFF has long warned that they are also high powered microphones parked above densely-populated city streets. Cities now have one more reason to follow the lead of many other municipalities and cancel their Flock contracts, before this new feature causes civil liberties harms to residents and headaches for cities.
In marketing materials, Flock has been touting new features to their Raven product—including the ability of the device to alert police based on sounds, including “distress.” The online ad for the product, which allows cities to apply for early access to the technology, shows the image of police getting an alert for “screaming.”
It’s unclear how this technology works. For acoustic gunshot detection, generally the microphones are looking for sounds that would signify gunshots (though in practice they often mistake car backfires or fireworks for gunshots). Flock needs to come forward now with an explanation of exactly how their new technology functions. It is unclear how these devices will interact with state “eavesdropping” laws that limit listening to or recording the private conversations that often take place in public.
Flock is no stranger to causing legal challenges for the cities and states that adopt their products. In Illinois, Flock was accused of violating state law by allowing Immigration and Customs Enforcement (ICE), a federal agency, access to license plate reader data taken within the state. That’s not all. In 2023, a North Carolina judge halted the installation of Flock cameras statewide for operating in the state without a license. When the city of Evanston, Illinois recently canceled its contract with Flock, it ordered the company to take down their license plate readers–only for Flock to mysteriously reinstall them a few days later. This city has now sent Flock a cease and desist order and in the meantime, has put black tape over the cameras. For some, the technology isn’t worth its mounting downsides. As one Illinois village trustee wrote while explaining his vote to cancel the city’s contract with Flock, “According to our own Civilian Police Oversight Commission, over 99% of Flock alerts do not result in any police action.”
Gunshot detection technology is dangerous enough as it is—police showing up to alerts they think are gunfire only to find children playing with fireworks is a recipe for innocent people to get hurt. This isn’t hypothetical: in Chicago a child really was shot at by police who thought they were responding to a shooting thanks to a ShotSpotter alert. Introducing a new feature that allows these pre-installed Raven microphones all over cities to begin listening for human voices in distress is likely to open up a whole new can of unforeseen legal, civil liberties, and even bodily safety consequences.
Originally published to EFF’s Deeplinks blog.
Filed Under: audio recordings, gunshot detection, microphones, raven, surveillance
Companies: flock, flock safety
Comments on “Flock’s Gunshot Detection Microphones Will Start Listening For Human Voices”
The potential for abuse is that if Raven picks up people talking to themselves about anything at all, the blackmail is going to happen
Two guesses at their response...
There are secondary risks
Police will of course abuse these, because that’s what they do with any technology. But that’s only the beginning.
When, not if, the supporting infrastructure for these microphones is hacked, the attackers will be able to monitor rather a lot of audio…so they will. And they’ll be able to feed to any of the numerous LLMs which are quite capable of replicating a given human voice given a sufficient sample size. They’ll then use that to synthesize whatever they want, in the voice they want, and inject it back into the system.
And that is going to be just as dangerous for the civilian population as the false alerts on gunshots. Police will hear what they expect to hear coming a place they expect to come from, they’ll show up, and nobody will be able to convince them that it’s a fake before they start shooting.
Re:
Never mind small children screaming at play. i am given to understand this is a thing.
Unintended consequences
Oh boy, is this going to get fun. I can’t wait for the stories of people messing with this system to start rolling in. “Help1 Help! I’m being attacked by people in furry suits!!” “I’m going to shoot them now!!” Rolls camera and waits for the military
Have they ever been to a school during lunch time? Kids are constantly screaming at the top of their lungs.
This is going to work exactly as well as their current offerings.
What could go (horribly, horribly) wrong?
Well I certainly can’t see any problems or tragedies waiting to happen(not to mention exploitation by bad faith actors) by having police scrambling across entire cities responding to false positives based upon random audio clips…
Kids playing vs. distress
I can’t speak for AI systems but I find sometimes the horse-play of children and adolescents can sound convincingly similar to distress, to the point I end up diverting my own course and activities to investigate. Typically, then I only to find that kids have gotten a bit rowdy in their fun and no actual harm is being done.
Since, for now, such systems tend to be less accurate than human sitch-assessment, I anticipate police are going to show up where kids are being kids, and will then find cause to be offended, the way resource officers can’t help but find reason to pugilize teens.
That is to say, this looks like Raven is a tool to justify more bullying of civilians by state agents.
I wonder if they can tell the difference between ...
live people and screaming on TV?