Imagine a pair of glasses that could tell a blind person who just walked into the room. And I don't mean through a phone app or through some clunky headset, just glasses. Ordinary-looking glasses that whisper, "That's your daughter. She's smiling." And suddenly the world becomes a little bit less dark for somebody who hasn't seen a face in years. Now, imagine those same glasses can read a menu in a foreign language and translate it in real time, can describe what's happening at a street crossing, can guide someone through they've never been in, narrating the turns, the stairs, the doors, all through a quiet voice that only they can hear. That technology exists. It's real, it's here, and it's built on genuinely impressive AI, real-time multimodal processing that takes in visual, audio, and spatial data, synthesizes it, and turns it into something useful before you've even blinking. As a computer scientist, this is the kind of application that makes me feel like the field is doing something that really matters. And this is precisely why Silicon Valley exists. To take a technology with the genuine potential to restore sight to the blind, hand it to a product management team, and watch them systematically optimize it until it is indistinguishable from a court-mandated electric ankle monitor.

The newer Ray-Ban Meta display models, which start at around a thousand dollars, add a private in-lens screen. Text, navigation, arrows, message previews, and AI responses appear directly in your field of vision, visible only to the wearer, which is perfect for when you want to look like you're staring deeply into the soul of your conversation partner, but you're actually reading the Wikipedia summary of the movie they're currently talking about. And the Meta neural band, which is a wristband included with the display models, reads electrical signals from the nerves in your wrist, allowing you to control the glasses through subtle hand gestures, scrolling through information, selecting options, interacting with the AI without anyone around you noticing that you're doing anything at all. Now, Meta's marketing department frames this as a super sleek telepathic interface straight out of a science fiction movie. In reality, standing in a public street attempting to dismiss a spam email by aggressively twitching your wrist at your thigh doesn't really make you look like Tony Stark. It makes you look like a highly stressed accountant who is secretly checking to see if their deodorant failed, or maybe a Victorian ghost trying to cast a very minor curse on a passing pigeon. It's a bold aesthetic choice is what I'm saying. For a sighted person, this is a convenience product. For a blind or visually impaired person, this is a different category of object entirely. Glasses that describe the room that you've just entered, that can read a label on a medicine bottle, that tell you when the traffic light has changed, that recognize your friend's face and whisper their names so you don't have to even ask, these are not incremental quality of life improvements. For the estimated 2.2 billion people globally who live with vision impairment, according to the World Health Organization, this technology could reshape daily independence in ways that previous assistive devices never could. That's the product at its very best. That's the version of the story that should have been the headline. Sadly, it wasn't.

Meta is still the embodiment of this is why we can't have nice things, and let me explain why. The culture collapse of Meta's smart glasses happened fast, and it happened for reasons that were entirely predictable to anyone who has been paying attention to how technology is actually used by real humans. Influencers, mostly men, started using the glasses to covertly film themselves approaching ladies in public and attempting to pick them up. The footage was posted online as content. The ladies being filmed did not know they were being recorded. The glasses look like ordinary Ray-Bans. That's the entire design premise. So, there was no visible camera, no obvious indicator that a recording was in progress. The recording LED exists, but it's small and users can have it disabled or covered by a modder for a price because nothing really screams, "I respect your personal boundaries." Like paying a guy on Reddit $50 to block out the only physical indicator that you're harvesting someone's face for content. It got even worse. Some wearers attempted to extort victims of covert recordings for money. A class action lawsuit was filed alleging that Meta misrepresented the glasses' privacy protections and the nickname stalker pervert glasses. The backlash became so intense that the product's own customers started leaving them at home. Danielle, a travel content creator, told Engadget, "A lot of men and their behaviors have ruined this product. I wouldn't feel comfortable around somebody wearing them, so I wouldn't expect anybody to be comfortable around me wearing them no matter where I am. At this point, they're like fancy paperweight." Will Kujawa, a freelance video producer, said the comments when he posted about buying a pair were so hostile he changed his mind entirely. I saw all these comments about if you wear those glasses, you're basically a predator or a creep, and I was like, "Oh, maybe it's not a good idea to have those." You know a consumer tech launch has gone spectacularly well when buying the product immediately signs you up for a community-led social trial as a suspected supervillain.

Then, New York became the first US state to ban smart glasses from all 1,240 courthouses, state, county, city, town, and village courts. Starting July 20th in celebration of the World Cup ending, anyone wearing them has to surrender the device to a uniformed court officer before entering. Even prescription smart glasses are banned. Bring a normal pair, but the trigger for this was genuinely perfect. When Meta CEO Mark Zuckerberg testified in a social media addiction trial back in February, members of his own team were wearing Meta Ray-Bans when they escorted him inside, prompting the judge to issue a warning about recording jurors. Meta's own security detail, wearing Meta's own product at Meta's own CEO's trial, inadvertently demonstrated why the product needed to be banned from courtrooms. You really have to admire the sheer unblinking consistency of the public relations strategy here, right? Walking into a massive court trial about the societal harms of your technology while actively wearing the exact surveillance equipment being litigated is a stroke of marketing genius usually reserved for cartoon villain syndicates. Is the corporate equivalent of turning up to a product liability lawsuit for a chainsaw manufacturer while revving a chainsaw in the gallery to show the judge just how smooth the idle is. Wisconsin and Pennsylvania already had partial bans. Illinois is considering banning them while driving. Royal Caribbean banned them from children's areas, restrooms, and medical facilities on their ships. And through all of this, Meta launched a $399 Kylie Jenner collaboration, smart glasses with generous AI voice built in, marketed as a fashion accessory. The protest collective, everyone hates Elon, responded by installing gorilla lenticular advertisements near Meta's London headquarters, holographic posters that flip between generous glamorous marketing photo and a skeletal x-ray vision reading we're always watching, referencing the 1988 film They Live. Their statement to HuffPost, "Meta and Ray-Ban's new AI glasses can be used to secretly record women and young people for sexual reasons. Simply put, that's abuse." They also added, "Billionaires could fund cures for cancer, so why are they funding glasses for perverts instead?" And then Lorde, performing at Madrid's Mad Cool Festival, told the crowd, "Increasingly in our world, it gets harder and harder to know what is real. Can I just say for the record, F the glasses. Don't get the glasses, not sexy." When a Grammy-winning indie pop icon publicly declares your flagship hardware to be an active threat to human sex appeal, it is time to close the laptop and order a dirty martini that tastes like a gin-infused Mediterranean sea. The product's cultural identity calcified around its worst use case before the best use cases had a chance to breathe.

And Meta's response to this has been, and I generally wish I were exaggerating, to try to make the glasses less detectable. Honestly, it just goes from bad to worse in the story. According to the Financial Times, Meta is testing a prototype of the glasses that continuously records audio and takes photos every few seconds without any warning light. Mark Zuckerberg has personally questioned whether the LED indicator should remain on during background AI mode, which is the mode in which the glasses passively observe everything you see and hear for hours at a time. The company is also exploring facial recognition capabilities. In practice, this would mean you walk into a room, you look at someone, the glasses identify them, pulling from Meta's database of 4 billion users, and display their name and potentially other information on the private in-lens screen that only you can see. You scroll through that information using subtle wrist movements via the neural band. The other person has no idea any of this is happening. There's no light, there's no sound, there is no visible camera, just somebody looking at you through what appear to be ordinary sunglasses while a trillion-dollar surveillance infrastructure silently catalogs who you are. It's the ultimate introverted superpower, provided your definition of introversion is covertly operating as a walking, breathing branch of the NSA. The guardrails on something like this would need to be insanely strong, not just strong, I mean literally architectural, built into the product at the hardware level, not bolted on as a software afterthought that can be modeled away for the price of a YouTube tutorial. Instead, Meta is exploring removing the one guardrail that currently exists, the LED light that tells the people around you that the camera is active. They'll try to sell it as some feature improvement, but realistically, it's a serious category change. It transforms the glasses from a camera you wear on your face into a surveillance device that is indistinguishable from ordinary eyewear. And this is being developed by the same company that contaminated a city's water supply through negligent discharge, that secretly paid hundreds of contractors to pose as teenagers to attack the competitors' chatbots, and that has a 66.8 failure rate on its own child safety systems. This is a safety record so profoundly unimpressive that you wouldn't trust them to operate a toaster, let alone manage a database of several billion human faces. To their credit, I guess, Meta recently announced that its glasses won't take photos or videos if the system detects the LED is covered, and that an upgrade will disable the camera if the LED has been physically tampered with or destroyed. That is a step, yes, but it's a step taken after the cultural damage was already done, and it only addresses the current product, not the next generation that may not have an LED at all. It's also remarkably optimistic piece of engineering. If you design a pair of glasses whose primary selling point to a specific demographic is that they can secretly record people and then you add a software patch that says, "Please don't cover the secret recording light." You haven't actually solved the problem. You've just issued a very polite request. This is the same as putting a wet paint sign next to a nuclear reactor and just kind of hoping for the best. Let's discuss for a second. There is a thought experiment that I keep turning over in my mind and I think it's worth sharing even though I don't have a clean answer to it yet. What happens if everyone has these glasses? Not just men with disposable income and tech enthusiasm. I mean legit everyone. Ladies, children, elderly people, people in every income bracket, every neighborhood, every country. If everyone is recording everyone all the time, does the power dynamic change? There is a version of that future that looks genuinely utopian. Universal accountability, permanent evidence of everything. Crime becomes harder because there's always a witness. Bad behavior in public carries consequences because there's always a record. The asymmetry that makes the current situation so toxic, people filming people without consent, disappears because everyone is filming everyone and everyone knows it. This is a theory of social harmony based on mutually assured destruction. The idea is that if we are all constantly and aggressively spying on each other, a beautiful polite utopia will just naturally emerge. It is the exact same logic that suggests the best way to prevent accidental kitchen fires is to keep your oven permanently turned up to 400° and filled with premium unleaded gasoline. Technically, everyone in the room becomes highly alert to the risks, but it isn't a particularly relaxing environment to make toast in. This is actually the premise of Dave Eggers' novel The Circle, later adapted into a film with Tom Hanks and Hermione Granger, where a tech company achieves total transparency by putting cameras everywhere. In the novel, this is framed as liberation. In practice, it functions as control. And the question the story keeps asking is who benefits from universal surveillance? The answer every single time is whoever controls the infrastructure, and in this case, that's Meta. All the data from all the glasses flows to Meta servers, processed by Meta's AI, stored according to Meta's policies, governed by Meta's terms of service. Universal surveillance sounds democratic until you realize that the democracy has a landlord, and the landlord is a company that parks super yachts in the harbors of cities where it just cut 1,400 jobs. A landlord that actively wants to know what you look like when you wake up, what brand of cereal you stare at, and exactly how many seconds you spend looking at a billboard before you feel the crushing weight of capitalism. And right now, the glasses are not universal. They're disproportionately owned by one demographic, and the people being filmed without consent are disproportionately in another. The power asymmetry is not theoretical. It is the bedrock of deploying this kind of tech. It's the reason the product is called pervert glasses and not accountability glasses. The surveillance argument, the idea that surveillance from below can check surveillance from above, only works if the surveillance is symmetrical. When it isn't, it's not accountability. It's just one more tool in the hands of people who already had too many. Could the technology equalize over time? Maybe. Glasses could become cheap enough for everyone. The accountability applications could become the primary identity of the product. Public trust could be rebuilt through transparent policy, strong guardrails, and a demonstrated commitment to preventing misuse, but that requires a company that treats privacy as foundational rather than as a speed bump on the way to engagement metrics. And Meta, based on everything I've covered on this channel, has several light-years before it gets that far. Yes, I know that's a measure of distance. I just wanted to see if you knew that as well. Throughout this video, I keep thinking about the blind person, the one who puts on a pair of glasses and hears their daughter's name for the first time in years because the AI recognized her face and whispered it through the speaker. That moment is worth building for. It's worth the research, the investment, the engineering hours. It's worth the company's full attention and its best intentions. And instead the world met this technology as pervert glasses. That's the first impression. That is what stuck. Not the blind person navigating a city with quiet confidence, not the tourist reading a menu in a language they don't speak, not the elderly person being reminded of a grandchild's name. The world met wearable AI as a tool for men to covertly film ladies on the street, and first impressions in technology, as in life, are almost impossible to undo. Meta had a choice. It could have launched this product with the accessibility story front and center. It could have partnered with organizations for the blind and visually impaired. It could have built the guardrail strong enough, hardware level, non-negotiable, not just a little software patch exposed, that the conversation would be about human augmentation rather than human exploitation. It could have earned the trust that a product this intimate, something you wear on your face, something that sees what you see, fundamentally requires. And instead we got a Kylie Jenner campaign, a prototype without a recording light, facial recognition that silently identifies strangers, and a cultural identity so damaged that the product's own customers are ashamed to wear it in public. The technology isn't dead, but its reputation might be, and reputations once lost are the hardest thing in technology to rebuild. Harder than models, harder than infrastructure, harder than anything you can solve with $135 billion in capital expenditure. You cannot advertise trust. You cannot celebrity endorse consent. You can only earn it slowly through behavior that consistently demonstrates you deserve it. Trust, as we all know, is a time series. Meta has not earned it, and until it does, the most transformative wearable AI product in history will remain in the public imagination a pair of pervert glasses. If you want to understand how Meta arrived at this point and why the most valuable advertising data set on Earth keeps being neglected in favor of decisions like this, I covered that in detail in the video that I'm linking here on your screen. Thanks so much for watching. I'll see you in the next one.