Meta’s Smart Glasses Can See Everything—But Should They?

Craig: meta smart glasses see everything, but should they? super sensing technology is an ai powered system that enables instant facial recognition, contextual reminders, and activity tracking. the system works by analyzing visual input through the built in cameras and processing this data through meta's advanced ai algorithm. the technology in these smart glasses is being deployed into two new smart glasses called aperol and bellini. these devices represent the next generation of meta's smart eyewear lineup. the company is also testing super sensing features on the current ray ban glasses ai models. what do you think about the super sensing features in these glasses that are a, looks like a supercharged version of what we've seen before when we covered the hack of the meta glasses, how people could go out, and it was scanning a new people wore. and we saw people go up to say, hey. i met you at the conference last year because it looked up post they made on social media. and yeah. so this becomes that but worse or pumped up or way better depending on

Jon: who you look at. because, yeah, more people can use it. but, this is the news that i've kinda been waiting for, but it also scares the crap out of me. so, i i love it because this would be great at networking events like we had talked about where, real time facial recognition, it can identify people by name as you encounter them. it can recognize objects, understand your environment, and trigger context aware reminders, such as suggesting you grab your keys or reminding you about your grocery list as you pass by a store, as you're leaving the house. right? that'd be that would be great for people with adhd. so the the the downfall about this is that they have to keep the cameras and sensors active at all times, which feed data consistently to the onboard ai. apparently, the core innovation is a hands free always on system, and that's where the ai acts like a real life assistant and anticipates your needs based on what it sees and hears and responds with voice commands, and responds to voice commands as well. like, if you're saying, hey, meta. start live ai or something like that. but the biggest hurdle, battery life. so tests that they're doing currently show that live ai feature currently only lasts about thirty minutes on the existing ray ban meta glasses because it's a high power draw, continuous sensing analysis, all that sort of stuff. so, i think you and i might have covered it or we at least talked about it. the new glasses that they were working on, that they were talking about probably maybe about six months ago, i feel like that's that's where this is gonna come into play, where it's gonna work a lot better. but i can't see it working more longer even with the new glasses, longer than, say, maybe two hours at the most. think?

Craig: i think it depends on the application. like, i think let's go through a couple of features, and then we'll talk about why i think we're gonna get longer running versions than we may think. super sensing enables real time identification without bystander indicators. like, you know, people don't know they're being identified. we have current battery constraints. environmental scanning suggests actions based on location and context. if facial recognition requires opt in but lacks bystander opt out. metastor's voice data for one year without an opt out option. continuous data collection risk creating public surveillance networks, interconnected wearables, earbuds, jewelry may expand sensing capabilities. nondisplay glasses, serve as a cost effective ai testing platform. so take a step back and say, okay. we are really creating a surveillance state, but, you know, that's why i think we're gonna see, police and military versions where yeah. the glasses are, like, hooked up to battery pack on the belt or on the backpack or whatever. so yeah. they'll get longer use. but, like, you know, you're doing events. you're doing security, i don't know, at a rally in the us, at a concert here, at whatever. you know, you gotta have one guy walking around with this, talking to the police database, so it's gotta be proprietary or whatever identifying people in the crowd. right? so you could really see that. but, you know, again, is that profiling? is that whatever if we're just you know, what are we detecting? we're using ai. anger? are we gonna detect these things? or are we just going to mhmm. read a database of criminals? like, what is, the long term intent of this? so that's one thing from a military or a police point of view. right? you know? yeah. it's easy now for something like this for a military group, insurgents doing whatever, to determine if their target is here in the square, find them quickly, you know, have the ai identified track and all that kind of stuff. so i could see that. but there's another thing to give the world of youtubers and tiktokers and sensationalists. we see this at the gym, at the grocery store, and at whatever. people are pranking on the subway and doing all this crap that this is a complete disruption to society, and now we're just gonna give them more tools to do that, as a civilian group of people. it just starts to scare me that that group will become a 100 time fold now that this will just become easy to do.

Jon: yeah. well, the thing is is, like, when you're out in a public space, you can't really expect privacy. and this just takes it one step further and makes it even worse. like, there's just there there there's gonna be no privacy. and, i think we're getting to the point where i don't know if you've, i don't know if you remember the movies where people are wearing specific clothing, specific things on their faces, and stuff like that. like, i think we're getting close to that. the the thing is how society manages these technologies, it's gonna depend on transparency, consent, and well designed privacy safeguards. so the tech itself could be game changing, but misuse or inadequate oversight could completely erode that. and that's what scares the crap out of me because we're so far behind when it comes to, oversight with ai right now, that i don't know if we're ever gonna catch up. i hope we do, but but we're so far behind because ai is moving so fast. i i don't know if if we're gonna properly deal with this, if that makes sense.

Craig: we're not. i think it's impossible. like, i don't understand how anybody could think we could deal with this properly. once you have technology like this that is contextual aware, it knows where you are, what environment this is. so not only, location, but the concert, the venue, all that stuff, and it's recognized everyone else around you, recommends task based on your environment, has smart reminders and object recognition. like, and, you know, you have hundreds of these people in the crowd. when i say these people, you know, glass holes, if you wanna call them that because to go way back, where are these glasses recording everything? yeah. and the battery technology is gonna get better. we know for a fact we talk about new battery technology all the time on the show, and then the stuff's coming out next year and the year after that. so it's gonna double, triple, quadruple the battery life of capabilities of some of these things. so now you can run these things for four or five hours and record, do local recognition, local recording, cloud recording, stream, and it's available to everyone all the time. i i feel like, we're in what was that where they had it wired in your head and you could feed the feed? yeah. where everyone was yeah. it just yeah. why do we need this? and i

Jon: like, if we're talking about end. if we're talking about current technology to kind of mitigate this, i think the quickest solution to kind of, try to control this or try to mitigate it as much as we can is geofencing, because they're always gonna be connected. so they're gonna know where they are. so if they're in a certain area, obviously, hackers are gonna get around that. but, but i think that that would be the best way right now is, like, set up a geofence. and if you're within the geofence, then your your your glasses don't work.

Craig: this has gotta be against every eu privacy policy i can possibly imagine.

Jon: yeah. maybe it's time to move to the eu.

Craig: well, i'm just saying that no. it's not really talked about here. like, they talk about the company response and the privacy issues, but we don't really talk about, hey. you know, what is the eu gonna do about this? but there are some extra insights on but, you know, they're making you can't do this. you can't do that privacy. everyone's, you know, data has to be controlled by them. so if you're walking around and it's identifying people and recording them and publishing it, all the time. it's one thing to randomly catch a picture. it's another thing in a public place. it's another thing to go in and out of public and private spaces, collect, video, and identify those people and, you know, do stuff with that data. that's completely separate

Jon: thing. yeah. agreed. yeah.

Craig: but having taken a step back, who doesn't want these glasses? who you know, you wanna walk in a room and identify the people you wanna need to talk to and be able to have their name pop up and whatnot when you go to talk to them and have whatever, maybe some social posts or your last email interactions be over there so you can have an idea?

Jon: yeah. yeah. your last conversation, have you had a conversation with them? like, with the amount of networking events that you and i are going to, i'm finding more and more of a need for something. and i'm sure i could do this just using, automation workflows, but, like, but i can't do it in real time. figuring out if i've talked to someone before, what i've talked to them about. there's so many people nowadays that it's tough for me to keep track of everything, which is why i've started going to less networking events.

Craig: well, you say you can't do that in real time, and i know we're getting a little out of the way. but if you had, like, siri or telegram or whatsapp or whatever, you can say, hey. where do i know bill hayden from or whatever? i just made up a name because you just met the person and have it go away, and it could come back, speak to you. it could do whatever. if you have an ear plug in, it could show you what have you. so, you know, it's not gonna be

Jon: yeah.

Craig: identify the person from across the room and do all that for you. you're gonna have to say, hey. who is this person?

Jon: yeah. well, gustav, definitely not siri because siri is the worst of them all. so, like, maybe gemini assistant or something

Craig: like that. you can hook siri now up to these external ais. you can use siri to call whatever you want. that's all i'm saying. yeah. i

Jon: see how siri has been the worst of all of them since day one, and they've never improved it. and, like like, when you ask siri a question, it still just gives you a web result. does it not?

Craig: yeah. it's it was it's worse than when they bought it. so yeah. it is what it is. but back to better glasses. let's not just be that's alright. i started the diversion. i see that there's gonna be a lot of legal frameworks that are lagging behind a wearable ai capabilities because i'm with you. what i would like is and this is probably what openai is coming up with because they're decided their io device, is not gonna be it's not a wearable, but it's something you would have on you that's context aware. so i think you're gonna have something that listens, either all the time or when you press a button, but just is with you, knows your geography, also knows all your agents and stuff in the back end, knows where you're supposed to be, your calendar. so everything is in context. yeah. so when i sit down with you, i don't have to tell it i'm sitting down with you. it records our conversation, and i go, i don't have to tell it anything. it knows all that. it transcribes it. it sends to my agents. it does all the things i'm supposed to do in the back end. i don't have to tell it anything. yeah. that, i think, is a future that they're heading for. but, again, that's just a guess. but you tie these meta glasses in the same type of thing, but now you're getting video and images. we're living in, a futuristic world. that's for sure. so there's one thing of recording everything, and then there's the other thing about being distracted by, feeds that you can now project on your glasses.

Jon: yeah.

Craig: so i think we're gonna be a far more distracted society and a far more you know, there is no such thing as privacy.

Jon: yeah. there was an article. i don't know if it was yesterday or the day before where mark zuckerberg said if people in the future, people that aren't wearing these ai glasses are gonna be left in the dust, which yes and no.

Craig: well, it what future? and i could see that. like, it's just like if you're not using ai agents, you're gonna be left in the dust. if you're not using workflow and automation tools, you're gonna be left in the dust. but when does the glasses become that next level of agent control?

Jon: yeah.

Craig: that's the you know, when you'll be left behind if you don't have your your alerts, your agents, you're controlling them all visually. you know what's happening. you can interact with stuff and say, hey. i like that. order me something from that restaurant. i'll pick it up in a bit. i see a theater up there. grab me some tickets. whatever. like, you know, there's so many different use cases, but having it context aware, being able to see what you can see here with knowing where you are, and just being able to say general things, and getting what you want back is gonna be an interesting world.

Jon: yeah. for

Craig: sure. but the world everyone's running around talking to themselves right now as it is because they're always on the phone. they're always distracted by screen. so i yeah. the more technology integrated we get, the less, human we become.

Jon: there's one thing i started doing that, this new, whatever, what i started doing because i used to walk around with my phone in my hand all the time. right? whenever i get up, i'd get in my hand and walk around even if i was just going upstairs. but now i've started trying to just leave it there. walk away. if i'm only going for a few minutes, like and it's it's been great. it's been good, a good exercise to do.

Craig: yeah. now we're gonna strap glasses to your head, and you're not gonna be able to see without them, so you won't be able to leave behind. yeah. anyway, i think the technology is really cool. i love the direction we're going. you throw this along with some of the night vision glasses slash, contact lenses we were talking about. you can really get some cool technology in these things. i'd love to know what you think. are you gonna be grabbing a pair and streaming this stuff online? are you gonna be using it for something else? i can see all sorts of, apps being, used for this. i think things are going off in my head right now. let's build apps for this today. what are you gonna use these? are you gonna avoid people who use these? or you think it's the end of privacy? let us know what you think in the comments below.