news

zuckerberg's new interview: smart glasses are the most perfect ai assistant and the ultimate digital social

2024-09-21

한어Русский языкEnglishFrançaisIndonesianSanskrit日本語DeutschPortuguêsΕλληνικάespañolItalianoSuomalainenLatina

editor’s note:
the meta connect 2024 conference will be held next week, and technology giant meta will bring their latest developments in artificial intelligence, augmented reality (ar), and virtual reality (vr).
among them, the new ar glasses orion, which were described by ceo mark zuckerberg as "impeccable", are also expected to make their debut at this conference.
this week, zuckerberg was a guest on the acquired podcast for a nearly 90-minute interview. in addition to some entrepreneurial stories, he also talked about his views on the future of ai, vr, and ar technologies.
this seemed to be a "side dish" specially prepared by zuckerberg before the "main course" of the meta connect conference. during the interview, he talked in detail about his ideas for "smart glasses" and future platforms, as well as the next focus of the meta platform.
the following are edited excerpts from the conversation:
smart glasses: the ultimate digital social experience and the ultimate ai avatar
q: please tell us the story of how (smart glasses) came about.
zuckerberg: at meta, we have been building social experiences for 20 years, initially in the form of websites and then mobile apps. but i never thought of us as a "social media company", we are a "social connection company", and what we are doing is building the future of human connection.
i think one of the important topics for our next chapter is,building what i think is the best experience, not just one built on someone else’s platform.
i think what you need is not a phone that makes you look down, which takes your attention away from the things and people around you. i think the ideal form is "glasses".
glasses can see what you see and hear what you hear, making them perfect ai assistants because they know what you’re doing.
beyond that, the glasses will be able to project holograms onto the world, so your social experiences with other people aren’t limited to those little interactions on your phone screen.
you can imagine that, in the distant future, we are having a conversation like this, but maybe one of us is not even there, like a hologram.
people like to intellectualize everything, but a lot of our experiences are very physical. with holograms and glasses, you can do things with another person in the physical world, and this sense of physical presence doesn't take you away from whatever you're doing.
i think this isthe ultimate digital social experience and the ultimate embodiment of artificial intelligence.
this is a huge project that we have been working on for 10 years. we need to create a new display stack for holographic displays and then miniaturize it to fit inside glasses.
beyond that, the glasses will need to be equipped with chips, microphones, speakers, cameras, eye tracking, and more to understand what the user is doing, and the battery will need to last a whole day.
we've been working on this for a while.we are very close to being able to show the first prototype machine we have, and i’m really excited about it.
we developed a lot of new technologies, but we also had to consider a lot of factors in terms of appearance, because the glasses had to look good. we worked with essilorluxottica, the manufacturer of ray-ban glasses, to explore what kind of functions could be packed into the glasses to make them as useful as possible.
ray-ban meta smart glasses
when we actually started making these glasses, i felt like this was almost a hands-on project for ultimate ar.
q: so when you first launched these (ray-ban) glasses, the chatgpt bot and the large model were not yet popular with the public, so these products were not initially built and launched as ai devices?
zuckerberg: yes, a few years ago i predicted that ar holograms would come before full ai, now i think it may be the other way around.
i remember calling the head of operations for the smart glasses and i was like, we should make this product better and make meta ai its main feature. then i went to their group the next week and they had a prototype and i was like, this is going to be a very successful product.
q: what do you think is the dna that runs through the company and enables it to continue to win?
zuckerberg: i think of us as a technology company focused on human connections, not a specific type of app, so we've never thought of ourselves as a website or a social network or anything like that.
for me,building glasses that allow people to feel like they are with another person no matter where they are is a natural continuation of the kind of applications we're building today.
open and closed source
q: i want to talk about the importance of open source and open source technology to you. i'm curious if you would agree that "meta has been the biggest beneficiary of modern open source technology."
zuckerberg: i think almost all the major tech companies right now are using primarily open source stacks, so, yes. i mean, we couldn't have built without open source.
i think this has been true for any new company since the 1990s, and open source has always been important and valuable to us.
q: i mean, you were the first big company to build on the lamp stack.
(lamp stack: an open source software platform commonly used to build dynamic websites and applications.)
zuckerberg: it's great and makes it easy to develop quickly and iterate quickly. our relationship with it is also interesting because we came after google, the first great company to build this distributed computing infrastructure, which is not an advantage for us because google already has it, so we might as well open it up. once it's open, the whole community will work around it.
we open up computing and make it an industry standard. now basically all other cloud service platforms are using open computing, and the supply chain has standardized around our design, which means greater supply and lower production costs. we save billions of dollars, and the quality of the products we use will also improve. so, it's a win-win.
we want to make sure we have access to leading ai models, i think, just as we want to build hardware so that we can build the best social experiences for the next 20 years.
i think we've been through too much with other platforms to be completely dependent on others, and we're a big enough company now that we don't have to be dependent on others.we can build our own core technology platform, whether it is ar glasses, mixed reality orai, so i think we have to do this.
but these things aren't monolithic like software, they're ecosystems, and they get better when other people use them, and that's great for us, too.
it also fits our position philosophically, and we had so many first-hand experiences where we tried to build things on mobile platforms and the platforms would ban us, which was very frustrating.
meta llama 3.1 open source model
q: who are your competitors?
tiekerberg:we face a lot of competition, there are a lot of aspects to what we do. i mean, we have social media competitors, we have platform competitors.
i thinkapple is a bigger competitor than people realizethey're also doing a lot of different things, but i don't know. i think for the next 10, 15 years, it's going to be an ideological battle.
what are the architectures of the next set of platforms going to be? are they going to be the closed, integrated, apple model that apple has always been? i mean, there are a lot of good ways to build things.
i think if you look at different generations of computers, pcs, mobile devices, they all have kind of closed integrated versions and open versions. the iphone has basically won, and i know there are more android phones out there, but the iphone is like the leader in smartphones.
i think there is a recency bias: maybe because almost everyone here has an iphone, there is a view that this is the better way (referring to closure). but i don't take it for granted.
in the pc era, windows is the leader in the open ecosystem. my goal for the next 10 to 15 years is to:build the next generation of open platforms and make them win.i think this will lead to a more vibrant tech industry.
this is not just a product competition, i think in some ways,it's like a very deeply values-driven and ideological competition around what the future of the tech industry should be and how open these platforms should be.
whether it's llama big model ai, smart glasses or something else, developers, especially individuals like me who are starting out in a dorm room, should be able to build awesome things without having to ask for permission.
about the future: virtual reality and ai
q: you spent a ton of money on reality labs (meta's vr division) and put so many people and resources into this project, which is definitely inappropriate. i'm kidding here. but i'm curious, why did you treat it so specially?
zuckerberg: when you get to the size we are now, you think about what will be important to do in the next 10 to 20 years? we are talking about how to create our own luck, and we can have a sense of how things will develop.
i'm sure,glasses and holographic images will become ubiquitous products.just like everyone had a cell phone before, and then they all switched to smartphones. if we can get everyone in the world who already has glasses to upgrade to glasses with ai, then this will be one of the most successful products in the history of the world, and i think it will go a lot further.
some early oculus (meta's early vr headset) users told me:“there’s a difference between creating good things and creating awesome things.”
good stuff is good, it’s useful, it’s something people use every day because it adds something to their lives. but awesome is different, awesome is uplifting and inspiring, and it only makes you more optimistic about the future.
so i think what we've done so far on social media has been pretty good, with over 3 billion people using it almost every day.
and now as i think about the next phase of the company, the next 15 years, i hope we can build more "great" things in addition to the good things, and i think they are all important.
so i think,a lot of the work we’re doing at reality lab falls into the “good” category.a lot of the work we are doing in ai will fall into this category.a lot of stuff in the app would go into this category as well, but i don't know.