Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror
AI Slashdot.org

Meta Is Adding AI To Its Ray-Ban Smart Glasses 23

Starting next month, Meta's Ray-Ban smart glasses will support multimodal AI features to perform translation, along with object, animal, and monument identification. The Verge reports: Users can activate the glasses' smart assistant by saying "Hey Meta," and then saying a prompt or asking a question. It will then respond through the speakers built into the frames. The NYT offers a glimpse at how well Meta's AI works when taking the glasses for a spin in a grocery store, while driving, at museums, and even at the zoo.

Although Meta's AI was able to correctly identify pets and artwork, it didn't get things right 100 percent of the time. The NYT found that the glasses struggled to identify zoo animals that were far away and behind cages. It also didn't properly identify an exotic fruit, called a cherimoya, after multiple tries. As for AI translations, the NYT found that the glasses support English, Spanish, Italian, French, and German.
This discussion has been archived. No new comments can be posted.

Meta Is Adding AI To Its Ray-Ban Smart Glasses

Comments Filter:
  • Yeah (Score:3, Informative)

    by skrugen ( 229044 ) on Thursday March 28, 2024 @07:53PM (#64352552)

    Wear those out in public and an NPC will murder you for them

    • by Press2ToContinue ( 2424598 ) on Friday March 29, 2024 @08:11AM (#64353188)
      The nicer illegals will just squat on your nose.
    • by Anonymous Coward on Friday March 29, 2024 @03:28PM (#64354316)

      You've clearly never seen these glasses.
      They look like completely normal glasses and can't tell there are electronics in them while being worn.

      Unless you're implying we now have mobs beating up people in mass for wearing prescription glasses to see.

      The one and only downside to these things is facebook. If any other company sold the hardware these are of the rare two I wouldn't mind trying out.

  • by fahrbot-bot ( 874524 ) on Thursday March 28, 2024 @07:58PM (#64352558)

    Meta Is Adding AI To Its Ray-Ban Smart Glasses

    Renaming them RAI-Ban would be paradoxical.

  • by serviscope_minor ( 664417 ) on Thursday March 28, 2024 @08:16PM (#64352578) Journal

    Not sure what having a mildly inaccurate object detector would do.

    You're looking at a pen! Or a cup!! And a car did you know you are looking at a car?

    • by dohzer ( 867770 ) on Thursday March 28, 2024 @08:54PM (#64352616)

      Much easier and cheaper to simply whip out Google Lens the one or two times a day you might need it.

      • by geekmux ( 1040042 ) on Thursday March 28, 2024 @11:06PM (#64352722)

        Much easier and cheaper to simply whip out Google Lens the one or two times a day you might need it.

        With massive advancements in navigation tech, we find many becoming incredibly dependent on GPS in order to simply navigate through daily life. Those that don’t even bother to memorize street names anymore, because that’s a map apps responsibility now.

        Now imagine what AI is going to do with general knowledge and learning. One or two times a day might describe the current generations used to learning the old fashioned way, but rest assured future generations will become incredibly dependent on it. They won’t even bother with learning much if the answer is provided by AI all day, every day.

        Human-powered learning will ultimately be viewed as inferior when AI proves to be the recognized authority. Sounds crazy? Who would you trust to get directions to drive 100 miles in an unknown area today? A fallible human, or your favorite mapping app? I thought so.

        • by account_deleted ( 4530225 ) on Friday March 29, 2024 @05:45AM (#64353040)
          Comment removed based on user account deletion
        • by nightflameauto ( 6607976 ) on Friday March 29, 2024 @09:25AM (#64353268)

          Much easier and cheaper to simply whip out Google Lens the one or two times a day you might need it.

          With massive advancements in navigation tech, we find many becoming incredibly dependent on GPS in order to simply navigate through daily life. Those that don’t even bother to memorize street names anymore, because that’s a map apps responsibility now.

          Now imagine what AI is going to do with general knowledge and learning. One or two times a day might describe the current generations used to learning the old fashioned way, but rest assured future generations will become incredibly dependent on it. They won’t even bother with learning much if the answer is provided by AI all day, every day.

          Human-powered learning will ultimately be viewed as inferior when AI proves to be the recognized authority. Sounds crazy? Who would you trust to get directions to drive 100 miles in an unknown area today? A fallible human, or your favorite mapping app? I thought so.

          Yeah, I can see this happening. There was a time I had about twenty-five or so phone numbers memorized and ready to dial up at any time because when you were out and about, you had to know the numbers you wanted to call, and you grabbed a payphone. Now? You just press the button to call Dalton, or Jenny, or Craig, and that's that. I force myself to dial important numbers every time just so that I always have them in mind. My wife, my mom, my dad, a couple friends I know I can count on if shit goes down. But most people I know don't know any number but their own anymore. They always rely on the phone.

          We're not that far away from blob existence, if we want it. They'll hook us to the machine to upload our basic body functions, and everything else will be dialed up on demand. Those with actual knowledge outside the machine will be weird, backwards, and probably ostracized for it. And society continues to "move on."

    • by martin-boundary ( 547041 ) on Friday March 29, 2024 @03:09AM (#64352924)
      How did you enable the adblocker on yours? This is what mine says:

      You're looking at a pen! Right now, Amazon has a special pen deal for $3.99 delivery in 90 minutes, enable 1-nod purchase? No actually it's a cup! Buy a bespoke tinfoil cup on Etsy for $210.50 plus delivery. Hurry while stocks last and nod to order. So sorry, this is actually a car you're looking at, definitely a car promise! Buy the new Xiaomi EV with complementary AAA battery charging pod before your friends get one! Just nod once up and down, or sideways to view extras Right Now! Don't Wait! Financing available if you touch the frame like you would when taking off your glasses by Ray-Ban(TM). Ray-Ban(TM) replacement lenses available from the Ray-Ban(TM) store, to go there now just dilate your pupils.

  • by ThosLives ( 686517 ) on Thursday March 28, 2024 @08:27PM (#64352588) Journal

    Isn't this more "adding a connection to the AI compute cluster" more accurate than "adding AI to the glasses"?

  • In, hmm, I think 1999 or 1998 I remember seeing professor Steve Mann of the U of Toronto (I think) gave a demo at Def Con via videoconferencing. I think it was so long ago it was "Cu-seeme". He had some kind of AI rig that he built himself. I remember it had an SGI machine (a heavily modified Indy, IIRC) which ran some kind of projection or LCD glasses. I can't remember how they worked. I just remember that he walked to the men's room and it blocked the Trojan condom advert on the wall and replaced it with an image of a waterfall in realtime. Then he showed how it could highlight objects in his environment (like items at a store on a shopping list). He then showed how it could see into other bandpasses of radiation as he could see the elevator coming up the shaft (kinda ghostly but you could tell) and the heat coming off the elevator call button. He also talked about and showed us some 3D models he could project over reality AI style.

    Professor Mann also answered questions from the crowd relayed by a devilishly handsome hacker. One of them was about the source code openness. The prof said he wouldn't even consider putting closed source in or on his body. I thought it was a cool answer.
  • by nospam007 ( 722110 ) * on Friday March 29, 2024 @05:52AM (#64353044)

    ' It also didn't properly identify an exotic fruit, called a cherimoya, after multiple tries. '

    I couldn't do that to save my life.

  • by cascadingstylesheet ( 140919 ) on Friday March 29, 2024 @07:08AM (#64353116) Journal

    It also didn't properly identify an exotic fruit, called a cherimoya, after multiple tries.

    Oh, well forget it then. Dealbreaker!

  • by awwshit ( 6214476 ) on Friday March 29, 2024 @09:37AM (#64353292)

    Good news! You do not have to feel a little slow and helpless anymore. Just put on our smart glasses with AI and be a Wizard.

  • by Iamthecheese ( 1264298 ) on Friday March 29, 2024 @10:32AM (#64353406)
    Acherimoya [wikipedia.org] A Cherimoya, also known as a custard apple, is a tropical fruit that belongs to the Annonaceae family. It is native to the Andean highlands of South America and is now cultivated in various tropical and subtropical regions around the world.

    The Cherimoya fruit is usually heart-shaped or round and has a green, scaly, and bumpy skin. The flesh of the fruit is creamy and custard-like, hence the nickname "custard apple." It is pale white in color and contains numerous black seeds that are inedible. Cherimoyas have a unique and pleasant flavor, often described as a combination of tropical fruits such as banana, pineapple, and mango. The texture is smooth and creamy, similar to that of a ripe avocado or a custard. To eat a Cherimoya, you can cut it in half and scoop out the flesh with a spoon, discarding the seeds. The fruit is best enjoyed when it is fully ripe, as unripe Cherimoyas can be quite acidic and astringent. Ripe Cherimoyas have a sweet taste and a fragrant aroma.

    Cherimoyas are not only delicious but also nutritious. They are a good source of vitamin C, vitamin B6, dietary fiber, and several minerals. However, they are also relatively high in calories and sugar, so it's best to consume them in moderation as part of a balanced diet.. t. AI.
  • by BigFire ( 13822 ) on Friday March 29, 2024 @11:20AM (#64353548)

    in search of a problem.

Been Transferred Lately?

Working...