Flock Safety's license-plate cameras now cover 40% of the United States, and a police officer in one town can query more than 100,000 of them in other states without a warrant.
Consumers hand over far more data than that to a phone or a smartwatch every day without complaining. The three people Kara Swisher put on a panel about the surveillance backlash all had the same explanation for why the cameras are the thing that finally broke: with a camera on a pole, there is nothing in it for you, and there is a city council meeting where you can shout about it.
"With the Flock camera, you know that there is no benefit to you. You're just being watched and it kind of makes you feel like a subject instead of the person in control."
The panel is one reporter who has been covering Flock since 2024 and whose work prompted a congressional inquiry, one academic who published on search-engine bias in 2012 and wrote the book on it, and one product reviewer who tests the wearables in question on her own body and negotiates with her editors about how much of her blood she will give up.
The full episode is covered here so you can skip it. 66 minutes of audio, 26 minutes of reading.
Here are the 15 arguments that matter.
👤 Guests: Jason Koebler, Co-founder of 404 Media, which has been reporting on Flock for the past few years; Safiya Umoja Noble, David O. Sears Presidential Endowed Chair of Social Sciences and Professor at UCLA, Director of the Center on Resilience & Digital Justice and author of "Algorithms of Oppression"; Victoria Song, Senior Reviewer at The Verge and author of the Optimizer newsletter, who covers wearables, health and fitness technology
🎙️ Host: Kara Swisher, Editor-at-Large at New York Magazine and co-founder of Recode
📰 Published: 14 September 2026 on YouTube (On with Kara Swisher)
🔴 YouTube | 🟣 Apple Podcasts | ⏱️ 66 min | ✅ Time saved: 40 min
Key Takeaways
Flock's cameras cover 40% of the United States and a local officer can search more than 100,000 of them with no warrant
Cities bought them to find stolen cars and missing children, then found police four states away using them
A Florida driver spent 13 days in jail because an officer searched the phrase "black SUV"
Her car had not been in the accident; she had driven past the scene
Flock paid for a study crediting it with solving 10% of all US crime, and Koebler says the researchers told him that is not what they found
Meta shipped the code for a face-recognition feature to millions of glasses and quietly pulled it back
The recording light can be disabled in software, drilled out, or simply washed out by daylight
Song's rule for wearables is that no device should make you the steward of a stranger's privacy
Her practical advice: if you are sensitive about health data, do not buy one at all
Noble would make these systems illegal until there has been a public vote on whether they exist
The panel's most actionable idea is a capital one: move pension money out of surveillance technology
University of California professors are running a campaign to do exactly that
All three think the backlash is real this time, and none of them thought so about Snowden in 2013
1. Why Flock, Not Your Phone
Swisher opened with the puzzle: people carry a surveillance device voluntarily and have for two decades, so why are the cameras the breaking point. Each of the three answered in turn, and their answers stack rather than compete.
Koebler's answer is that Flock is a target you can actually reach. "I think that Flock has become this avatar for people to push back against surveillance that they know that they're subjected to."
The consumer products come with a trade and with no realistic opt-out. The cameras arrive in your neighborhood with neither, and there is a venue — the city council meeting — where an ordinary person can object. Pushing back against Meta or Google alone is far harder
Swisher drew the parallel to data centers, where the objection to AI has become physical because the buildings stand in communities
Noble's answer is about where the cameras go. They are disproportionately installed around cities, which she reads as a signal about who the target is. The pitch is security, built on broken-windows theory, and the practice is different: "We actually see that these are used to ensnare and entrap people"
Her framing is continuity, not novelty — the same overpolicing of Black, brown, working-class and poor communities, now permanent and hard to evade
Song's answer is the missing trade. A smartwatch gives you health data, glasses give you footage of your children, and the camera gives you nothing
She added a second change: who is behind the camera. Trust in technology has fallen, and the AI backlash, the device backlash and the surveillance backlash are combining rather than running separately
2. A Network in 40% of the US
Koebler explained the product, and the key fact is that the cameras are not the innovation.
Automated license plate readers have existed for more than a decade. What Flock did was network them together using artificial intelligence, which he called the company's differentiating factor
Each camera photographs a passing car and adds it to a queryable database, and "there's no warrant requirement for that"
"Right now, Flock is in 40% of the United States." Where a department used to search only its own town's cameras, officers can now search thousands across a state or more than 100,000 nationally
The mismatch is that local needs are not national needs. Cities signed up expecting missing children and stolen cars: "And they woke up one day and realized that it was being used by cops four states over to search for undocumented immigrants or to search for a woman who had an abortion or by cops to stalk exes, things like this."
Retention was 30 days, and Flock has since moved to recommending a seven-day period, which Koebler believes is a recommendation rather than a requirement. Swisher pressed the obvious limits: material can be copied, and the database can be queried repeatedly over time, which Koebler confirmed
3. The Guardrails Came Later
The sales motion explains the backlash better than the technology does, and Koebler was explicit that the sequence is the problem.
Flock built the network town by town, and many deployments were pitched as pilots or bought with police discretionary funds, which meant no formal city-council approval
The company has since retrofitted controls — shorter retention, and automatic auditing for anomalous searches, such as one officer querying the same plate repeatedly, which may mean he is stalking someone
"What happened was the company didn't anticipate these problems ahead of time." Swisher's gloss was that the company is fixing it now while asking why anyone is angry, and compared it to Facebook's standing defense that it did not know people would do that
Koebler's wider objection is about process, not features. "I think for me the big issue with this is that this technology network has been built out by a private company without the sort of robust public discussion that we're having right now."
He put it in the same category as data centers: non-disclosure agreements, build first, and ask forgiveness rather than permission
4. Ring, Apple and the Pivot
Swisher noted that Amazon's Ring ended its partnership with Flock earlier this year and is updating its technology to stop third parties viewing customers' footage without consent, and asked Song whether that is a response to Flock specifically.
Song's read is that the whole industry is reacting at once. She pointed to Apple leading its own event with privacy, tied to new Apple Watch audio intelligence features, and read that as an answer to the Flock backlash and to what the public has been calling Meta's glasses
Her explanation of why it is happening now is about AI's appetite. Companies are hungry for data from every device because the product only works with it
The result is a merged anxiety: "AI is coming for my job. AI is also coming for my personal data. AI is coming for every aspect of my life."
Her verdict on the reassurance campaign was blunt — the companies have an incentive to lull customers into believing they care about privacy, and it is a little too little, too late
Swisher's correction, for the record: "By the way, the technical term is creeper glasses, not pervert glasses."
5. Predictive Policing's Past
Swisher read Flock's own mission statement — to eliminate crime with a principled framework that protects privacy and promotes objectivity — and asked Noble what she made of it.
Her answer starts with what the systems are trained on. The underlying logic is using historical data to decide where the hot spots are, and the historical data records who was policed rather than who offended
She placed policing itself in that history, tracing its invention in the United States to the monitoring of enslaved Black people
The technology predates Flock and Ring. She named PredPol as the earlier generation of software making the same prediction, and connected it to racial discrimination, redlining and containment
The circularity is the mechanism, and she stated it plainly: "So if you have more arrests in certain communities then you're going to have algorithms that are going to predict to go there to do more policing and more watching."
Swisher made the same point in one line: more Black people are in jail, therefore they must be more criminal — which is not true, they are more arrested, and that is different
6. Condor Cameras Track People
Flock has spent recent weeks insisting it photographs plates, not people. Koebler's answer was that both halves of the sentence are wrong.
The logical objection first. "We're not tracking people, which of course if you're tracking cars, you are inherently tracking people."
The product objection is bigger. Flock OS lets police pipe in footage from any surveillance camera they have, whether Flock's or not
The Condor cameras are pan-tilt-zoom video cameras, not still cameras. Some were exposed to the open internet late last year, and Koebler went and stood in front of them in Bakersfield, California — they detected him and zoomed in while he walked his dog across the street
They are not only on roads. He listed bike paths in suburban Atlanta, street fairs, pools, community centers and pedestrian walkways, all feeding a system police monitor in real time
His conclusion is about ambition rather than capability: the company wants to look at everything
7. The 13 Days in Jail
The panel's strongest objection is not that surveillance is intrusive but that it is wrong often enough to jail people, and one case carried it.
Song rejected the premise of the trade-off itself. "And so this whole premise of that you're trading off your privacy for your security and your safety is based on the idea that these systems are infallible and accurate and it's always going to catch the right person. And that's just not true."
The case: a woman held 13 days despite having evidence she was innocent, because the system said she was the one
Koebler supplied the mechanism, and it is worse than a bad match. 404 Media reported this summer on a Flock feature called Free Form, which searches the plate-reader images by description rather than by plate. "What happened there was the cops searched for a black SUV."
Her car was flagged because she had been near the site of a hit-and-run that killed three people, and an officer arrested her at her house on manslaughter charges. "Her car was not in any accident whatsoever. She was not involved. She just happened to be nearby."
Free Form searches anything visible, not just vehicle types: a protest sign, a bumper sticker, a dog. If a dog bite is reported, an officer can search every dog that passed a camera
Koebler dismissed the "motion activated" defense. "Another thing that Flock likes to bring up is that these cameras are not always recording. They are motion activated." Motion means any person or car going past
Noble added the accuracy research. Facial-recognition systems are least accurate on Black women's faces and most accurate on white male faces, and people have been wrongly incarcerated as a result. "So this idea that we can scientifically somehow solve these problems through surveillance is really untrue."
Swisher noted for the record that Flock has told 404 Media its products have no facial recognition and none in development
Noble's counter-question is where the money goes: "And so I think that the question is really about the underlying causes the root causes rather than making an incredible amount of money off of drones and flock cameras using the pretext of crime to do it."
A feature-velocity problem sits underneath all of it. Koebler's point is that a city signs one contract and then receives every new capability the company ships, with officers experimenting on it
8. A Battle of the Anecdotes
Swisher cited local reporting that Atlanta, where Flock is based, is among the most surveilled cities in America while the share of crimes considered solved has not meaningfully risen, and asked how efficacy should be judged at all.
Koebler's honest answer is that nobody can judge it from the outside. Each side trades anecdotes: Flock cameras used to surveil immigrants at the behest of ICE and a woman who had a self-administered abortion, against a missing child found and a murderer caught
Crime has too many causes for the attribution to work. Socioeconomic factors decide whether a crime happens and whether it is solved, which makes single-case evidence close to worthless in both directions
The study is the sharpest example. Flock, he said, "paid for a study a few years ago that said that they were responsible for solving 10% of all crime in the United States, which is an absurd claim." He then went to the researchers: "And also I spoke to the researchers who did that and said this is not what we actually said."
The rhetorical counterattack is the part he objects to most. Now that the public debate is happening — on social media and at every city council meeting — the companies and the police answer it by asking whether the objector loves criminals
9. How Bias Enters the Index
Swisher asked Noble to explain, for a non-technical listener, how stereotypes get into training data, naming the research of Timnit Gebru and Joy Buolamwini alongside Noble's own.
Noble has been on this since 2010 and published from 2012, when the word "algorithm" was not in ordinary usage
The mechanism she described is ranking by link, and links can be bought. A search engine treats a page as credible when enough pages link to it, and the network of pages that do the linking can be optimized by whoever has the most money
Her own finding is the example. "So in the case of the study that I did, I found that when you did queries on black women, black girls, Latina girls, Asian girls, the majority of the representation, misrepresentation of them was pornography." The pornography industry could optimize those keywords better than the women and girls being described could
The second failure is the absence of any standard of evidence. "So, if you are indexing the internet and you have no standard or value by which to evaluate, let's say, evidence-based research against a subreddit that's, a neo-Nazi thread, or a misogynist thread or something in the manosphere, you actually put those things all up equal to each other."
She gave the same-era example of searching for Jews and getting Holocaust-denial and antisemitic sites in the first few results
The third layer is the story told about the output — that the ranking is neutral and objective, which is how people come to relate to it
Swisher's matching anecdote is an argument with Larry Page, who told her all news is the same. Her objection was that the New York Times is not the National Enquirer, and that treating them equally is itself a weighting in the Enquirer's favor
10. Meta's Name Tag Problem
Song had recently written that Meta was playing its cards all wrong on smart glasses, and she walked through why.
The feature was revealed rather than announced. A New York Times investigation surfaced an internal Meta document discussing rolling out a facial-recognition feature at a moment when privacy advocates were distracted by political turmoil
404 Media's reporting made the fear concrete by spotting Meta glasses on Customs and Border Protection agents, which turned the question into whether the government could identify people through them
Meta's response was that it is a closed system on Instagram, based on the user's own contacts — a clarification that did not travel
Then Wired reported that Meta had shipped the code for Name Tag to millions of devices and quietly walked it back, which is what turned the story into a backlash
The precedent is not hypothetical. College students modified a pair of Meta glasses a couple of years ago to dox people instantly against existing facial-recognition databases
Song's objection is the everyday case, not the espionage one. "It's just creepy to have the idea, especially if you're a woman, that you're walking down the street and a pickup artist could be looking at you and finding out your name and your address, pretending that they know you."
The recording indicator does not work. Meta's privacy policy amounted to telling users not to be a jerk, and the single LED can be defeated in software or by drilling it out. "It's very easily washed out by bright sunlight." In her own testing she has asked people whether they can see the light, and they cannot
The reassurance about battery life is temporary. "Well, the reality is that these glasses only last about 30 minutes of continuous recording." Meta's stated ambition, per reporting she cited, is devices that sense or record continuously
11. The Rules They Would Set
The show's outside-expert segment put a question to all three: what should companies never be allowed to collect, and what single rule should exist before this infrastructure becomes impossible to unwind. The questioner disclosed that she is a Flock customer whose street installed cameras after a break-in.
Noble answered from the same position. After vandalism at her home, her family chose a closed system that let them see what was happening without feeding police and without risking a false arrest of her teenage son. "I think I thought that was the more ethical personal choice for our family."
Her principle: installing surveillance for your own safety does not entitle you to surveil the neighborhood, and the alternative she prefers is the neighborhood listserv and people looking out for each other
Her one rule is a moratorium: "I think these systems should be illegal until we have done the proper vetting had the proper public discourse and we have more democratic participation in whether they exist in our lives at all"
Song's rule is about consent and about who carries the burden. "I don't think that anyone should ever be put in the position to be the steward of someone else's privacy given these devices." Being made responsible for a stranger's privacy because you share a space with them is what has turned her off the convenience argument
Her second rule is that the decision cannot be left to the companies. "The number one rule I think is that we cannot leave it to private companies to make these decisions." Her worked example: if smart glasses become someone's primary seeing device and a courtroom bans them, that person is left without sight unless they carry spares — a question for legislators, not product managers
On the accessibility argument she was sharper still. Swisher noted the pitch that the glasses help veterans and blind people. Song: "Those people too deserve privacy. Those people too deserve a higher standard."
Koebler's first answer is that facial recognition should be a non-starter, and his broader one is about automation: a human used to review footage after the fact, and now an AI watches continuously across a far larger network
"I think that we are marching towards a place where it is very difficult to exist in public without running a foul of some local ordinance or small crime."
Swisher named Larry Ellison as the person who has argued the surveillance is good because people then behave, and Koebler agreed that is the goal some of them imagine
His concrete ask is a warrant requirement, against the industry's claim that it would make the technology unusable. He cited studies putting the average warrant at roughly three minutes to process, and argued the friction alone would prevent a lot of abuse. Companies ask to be regulated and then go quiet when a warrant requirement is proposed: "It appears to be a non-starter for Flock and companies like it."
12. Wearables Want Your Blood
Swisher turned to consumer devices, citing an Electronic Frontier Foundation finding that most wearables lack basic transparency reports and key privacy features, and an estimate that nearly half of Americans own one.
Song described the data pipeline. A smartwatch samples heart rate and movement continuously, feeds it into proprietary algorithms that differ by company, and increasingly into an AI that returns personalized insights, with personalized medicine as the promised destination
The appetite keeps widening. Companies are tying up with continuous glucose monitors to get at metabolism, want hormone data, ask users to urinate into a cup, and "They want to measure your sweat through your feet for a scale."
The stated benefit is a complete picture of your health; the commercial use is advertising — the same structure as Flock promising to solve crime
Her warning is that the category has changed. These are no longer health trackers but devices that will be on your body listening
Her advice to readers is unusually direct for a reviewer. "And I always say that if you're very sensitive to your private data, your private health data, you do not want a wearable device. I say this as the wearable lady."
The reason is the chain of agreements: the device has terms, the phone has separate terms, and every third-party connection such as Strava adds another, possibly weaker, set
"HIPAA exists but it's woefully inadequate and has been outdated for wearable technology since its inception."
Who the devices are for is a short list. People with a condition to monitor, the chronically ill, the hard of hearing and the low-vision, where the trade buys real independence
"But if you are an average person, sometimes the healthiest thing you can do is not engage with these devices." Wearables are for long-term trends, and can be ignored most of the time
She sets her own limits and negotiates them at work. She tests continuous glucose monitors because of a family history of diabetes, and has agreed with her editors what level of invasiveness she will accept — an increasing number of companies now want her blood
Swisher's own version: Zeke Emanuel told her during a CNN series that when you have not slept well, you do not need a ring to tell you
13. Baby Tech as Snake Oil
Swisher raised New York Times reporting on Nanit and other smart baby monitors selling benefits such as optimized nap predictions, and the Times' figure that the wearables market is approaching $100 billion.
Noble's judgment on the category is unqualified: much of this wearable technology is snake oil, and people do not understand that
Her evidence is a hospital-grade device failing. During COVID the pulse oximeter could not accurately read oxygen levels in African-American or more melanated people
"So imagine putting the faulty science on a newborn baby." "I'm going to tell you that there's not a lot of research on newborn babies", and clinical research in medicine is mostly conducted on white men
Consent is the second objection. A baby cannot agree to being enrolled, and many parents have spent the past decade taking their children off the internet, partly because of how children's images are used
The marketing is what she objects to most. "People want to protect their babies. They want to believe that they're giving their children the very best." Using that to sell a monitor makes parents and caregivers the objects of manipulation
Swisher's response is the episode's best joke and also an argument. "you cannot optimize naps", and babies are not committing many crimes
Her own use of the technology runs the other way. She allowed a fall camera for her mother, who was falling often, and insisted on an on-off switch. She also used an Apple Watch when she had a stroke, and noted the devices can help doctors detect conditions such as atrial fibrillation — while that utility depends on ever more detailed data
14. Move the Pension Money
Swisher asked what arguments companies use to get people to hand over data, and how good those arguments are.
Noble's answer is that the persuasion has moved from the consumer to the employer. The public is deeply suspicious of generative AI, and then goes to work — especially in white-collar jobs — and is required to use it. Enterprise engagement is the smart play
Her second example is pricing, which the panel had not touched. Surge and predictive pricing mean the grocery bill differs by shopper: "I'm paying more than Victoria is paying for the same bag of carrots."
The third route is fun. Products are gamified and sold on not being left behind, with the friend.com necklace — advertised across Los Angeles — as her example of a device that records everything you do and comments on your life
She credited the reporting. The Verge and 404 Media doing this work is how the public understands what the stakes are
Her practical instruction is financial, and it is the most actionable thing anyone said. "One of the most powerful things that people can do if you have a pension or you have a 401k or you have any way to direct your money out of surveillance tech that is something you can and should do."
"We're doing that as professors right now in the UC." She named CalPERS and CalSTRS as the funds that should be pressured, on the reasoning that pension capital is where these companies get the money
"Our instincts in the public that these things are not what we want are the right instincts."
Swisher's own line on the bargain: the companies say the product gets better if you give them your data, and that is their business problem to solve rather than a reason to take from you
15. Is This a Movement?
Swisher closed by asking whether the backlash can produce systemic change, said she doubted it, and quoted a European regulator who told her people are trading convenience for control of their lives.
Koebler thinks it is different this time, and he has a baseline. He became a technology journalist just before the Edward Snowden revelations of 2013, which changed very little in privacy law
What is new is the volume and the literacy. "I absolutely do think that there's a movement there like you identified." Videos about it on Instagram are informed and have 10 million views rather than 10,000
And it is producing outcomes. "You're seeing city councils cancel these contracts and you're also seeing attempts by the people who are opposing flock to make this a more universal discussion about surveillance technologies in general."
He was careful about the other side. There is a great deal of money riding on the technology, and whether politicians listening turns into meaningful change or into cosmetic packaging remains to be seen. "I'm feeling a bit more optimistic than I have in quite some time just because of the amount of energy that we're seeing right now."
Song's evidence is that the public has killed this product category once already. Google Glass was defeated by social pressure, and the same language is showing up around Meta glasses
She does not endorse the method — her readers talk about ripping them off people's faces, which she warns against — but the effect is real
Her colleague Mia Sato reported on retail workers being harassed by people wearing them, and the worker's answer was the elegant one: say "Meta, stop recording" out loud
"So people can do things to make these devices unattractive and that will force companies to either adopt a less evil way of implementing that." Apple's privacy framing around the new audio features is the response she expects
"So I do have hope there cuz we killed smart glasses once." She remains a skeptic that smart glasses are inevitable at all
Noble is hopeful for structural reasons rather than cultural ones. Unions are now bargaining over the use of AI in jobs, which she wants pushed harder
"I mean, nation states are talking about we need to get our governments and our citizens off of Microsoft." She described a European divestment from US big technology
The fringe alternatives are moving toward the mainstream: a renaissance for Linux, home servers, keeping things local, mesh networks and Bluetooth networking
Her closing warning is about who assembles the profile. As people learn what it means for a company like Palantir to be the infrastructure where all the data converges into profiles that determine how a person can move, the comparison to China's social credit system gets uncomfortable: "We should be worried about what the United States is doing."
Her list of places to act: move the money, bargain at work, and confront school boards about predictive AI in children's lives
Bonus Insights
Swisher's framing of the whole subject came in the first minute: the question is what is being collected about us and who we are in the matrix. "FYI, you're the battery."
The Google Glass story she tells is a decade old and still the cleanest illustration of the social problem. "A true story, I was at a Vanity Fair Oscar party and Sergey Brin walked in with Google Glass and nobody would talk to him." He asked her why, at a party full of celebrities who did not want to be filmed
A live example arrived from her softball team. A teammate wore a pair, hit a home run and turned them on: "He hit a home run and he turned them on and I was like oh my god I'm being recorded. I didn't even know and I sort of don't know how to react in that moment."
Swisher has acted on her own advice. She no longer wears an Oura ring, has moved to an analog watch, and has taken the connected devices off her house — while conceding the phone is still with her all the time
Her Ring objection is about third parties who never agreed. Your household may be fine with the camera; the delivery driver and the gardener are being opted in
Apple got one point of credit from the panel, for making the new listening feature opt-in rather than on by default, which Swisher said a Facebook version would not have been
Song's aside on the Apple announcement was less generous — she noted the same week's renaming of Lake Ontario to Lake America, and told listeners to turn the always-listening feature off
Koebler named the mundane version of the trade-off in TSA PreCheck and Clear, which people accept to save a few minutes, and said the discomfort starts when you no longer know where the data flows
The company he would trust most is Apple, not for a perfect record but for on-device processing and user control over where data goes
The panel's bottom line is that the surveillance argument has stopped being about whether a camera is intrusive and become an argument about accuracy, consent and capital: the systems are wrong often enough to jail innocent people, nobody in the frame consented, and the most concrete thing an ordinary person can do about it is take their retirement money out of the companies building them.
Products, Companies & Tools Mentioned
Flock Safety (License-plate readers in 40% of the US, queryable without a warrant, plus Flock OS, the Condor pan-tilt-zoom cameras and the Free Form description search)
Ring (Amazon's camera business, which ended its Flock partnership this year and is limiting third-party access to customer footage)
Meta (The Ray-Ban glasses and the Name Tag facial-recognition feature whose code Wired reported was shipped to millions of devices and quietly withdrawn)
Apple (Led its event with privacy, shipped opt-in audio intelligence on the Watch, and is the one company Koebler named as at least nodding to privacy)
Google (Both the search ranking Noble studied and Google Glass, the product the public killed a decade ago)
PredPol (The earlier generation of predictive-policing software Noble put in the same lineage as Flock)
Oura (The ring Swisher has stopped wearing, and the device Zeke Emanuel told her cannot tell you anything you do not already know about your sleep)
Strava (Song's example of the third-party connection that adds a separate, possibly weaker privacy policy to your health data)
Nanit (The smart baby monitor selling optimized nap predictions, in a wearables market the Times puts near $100 billion)
Friend (The wearable necklace advertised across Los Angeles that records everything you do and comments on your life)
Palantir (Noble's example of the infrastructure where the data converges into profiles that decide how a person can move)
Microsoft (The company European governments are discussing getting their citizens off, in the divestment Noble described)
404 Media and The Verge (The two outlets Noble credited with making the stakes legible; Koebler co-founded the first and Song reviews for the second)
CalPERS and CalSTRS (The pension funds Noble wants pressured, as part of a University of California campaign to move retirement money out of surveillance technology)
Books & Resources Mentioned
Algorithms of Oppression – Safiya Umoja Noble (Her book on search engines and racism, and the source of the study finding pornography dominated results for queries about Black, Latina and Asian girls)
Gender Shades – Joy Buolamwini and Timnit Gebru (The facial-recognition bias research Swisher named alongside Noble's own work)
404 Media's Flock reporting (The coverage since 2024 that prompted a congressional inquiry, and the source of the Free Form and Customs and Border Protection stories)
EFF's two-part doxxing guide – part two (Reducing your footprint, and what to do once you are being doxxed — the guide Swisher raised in her closing question)
Surveillance Self-Defense – Electronic Frontier Foundation (The EFF's standing guide, from the organization whose wearables report Swisher cited on missing transparency and privacy features)
If this was worth your time, send it to someone closer to the industry than you are.
Get the latest market chatter as it happens: