What happens to Bose when headphones become AI?

图片

Today, I’m talking with Lila Snyder, who is the CEO of Bose.

You certainly know Bose — it’s one of the most famous brands in all of consumer tech. The company started 60 years ago selling speakers to consumers, and its focus on research and development has led it to be a leader in both car audio and noise-canceling headphones.

Lila was last on the show about four years ago, and since then she’s dramatically rearchitected the company. The focus on research is still there, but she’s sold off Bose Professional, acquired the high-end audio brands McIntosh Labs and Sonus faber, and, perhaps most importantly, started licensing all that core Bose technology to other companies.

图片

Verge subscribers, don’t forget you get exclusive access to ad-free Decoder wherever you get your podcasts. Head here. Not a subscriber? You can sign up here.

That means that while you can still go buy a Bose soundbar or Bose headphones, the company is also turning into a B2B business, and you can get Bose technology in Skullcandy headphones and Epson projectors, among many other products.

That big change in the business comes with a big change in structure. Becoming a software company in addition to a product company comes with a whole new set of challenges, and Lila and I talked about how it’s all working out.

I also wanted to know if this is all a reaction to the potential rise of AI wearables — products that might end up looking a lot like headphones, and which need all kinds of audio processing to enable voice input.

Is Bose just trying to sell tech to a range of new companies that might disrupt headphones entirely? Or does Lila have a plan for a headphones market that was already disrupted when phone vendors like Apple and Samsung made their own products like AirPods and Galaxy Buds work better than the competition?

We got deep on this, and Lila was pretty open about the challenges and the opportunities here. As Decoder listeners know, I’m a huge audio nerd, so I loved every minute of this conversation. I hope you do too.

Okay: Bose CEO Lila Snyder. Here we go.

This interview has been lightly edited for length and clarity.

Lila Snyder, you’re CEO of Bose. Welcome back to *Decoder*.**

Thank you so much.

I am really excited to talk to you. It’s been four years since you were last on the show, which seems entirely too long.

A lot of things have happened. Bose has bought some companies. You’ve sold some divisions. The idea of what headphones are is largely the same in one important way, but maybe going to be very different in another important way. How are you thinking about Bose today?

We are having a lot of fun in what I would describe as our second chapter of the transformation of Bose since I joined. In some ways, many things are the same. We’re still making amazing products for our customers. We’re still excited about bringing that new technology forward in both our consumer business and our automotive business. But I would say two things fundamentally have changed as we’ve thought about growth for Bose for the future.

For 60 years, it was all Bose, and we’re now a multi-brand company. So as you said, we’ve made a couple of acquisitions. One of those was the McIntosh brand, a legendary 75-year-old audiophile brand that we absolutely love, and also Sonus faber, which is an Italian craftsmanship heritage brand that’s also been around for over 40 years. And so that brought us two things: from one brand to multiple brands, but also into the luxury audio space.

The second change for us is that we’ve always been an amazing products company where we developed technology through research and a lot of development cycles, and all of that technology went into Bose products. And today what’s different is we’re still doing that, and we’ve added a technology business we call Audio Tech, where we’re taking that technology and starting to put it into other places.

The big insight for us was that sound is really important. It’s getting more important, we think, given everything that’s going on in the world around us. There are so many products and devices out there that are using sound — whether that’s audio, voice, or even noise-canceling, where those things are incredibly relevant — but they’re not products that Bose would make. They’re not targeted at our customer set. But we really have a strong belief in bringing great audio to as many places as we can.

The last time we talked, you were in the middle of becoming a more functional company. In particular, you had said the last step was the engineering organization, which is really the heart of Bose. It had been very divisional, very organized into products. You were centralizing that.

Having an audio tech division that sells core technology feels like a very natural offshoot of that. “We’re going to have a big central engineering team. It’s going to feed into our products and other products,” and then you can abstract that a little bit and sell it to other people. Who are the customers for the audio tech division?

So it’s a B2B2C business, if you will. So our customers are large companies that are putting products into the market. So I’ll give you a couple of examples. We’ve partnered with Motorola on a series of products from earbuds to making the phone itself sound better.

So Motorola is a good example. We’re partnered with Xreal. They’re making pretty awesome XR glasses, right? So that’s an experience where sound really matters a lot. We’ve had a lot of fun partnering with them. You’re starting to see Bose technology show up in a lot of different verticals, from mobile to compute, to other people’s headphones and audio products for sure.

We just released, with a partner called Sena, a motorcycle helmet communications device. And we’re excited about all the places we’re bringing our technology. It’s actually creating a broader aperture for how we think about invention and research at Bose as well, which is really exciting.

I want to come to that because that seems like a big change. And the reason I asked about customers is that having a new set of customers really changes the focus of the company. A lot of us are familiar with audio brands in particular partnering with device makers, partnering with automakers.

Yeah.

Maybe it’s just a logo. I think you have been very proud that Bose has not been just a logo in many places, but for a lot of people it’s just a logo. What’s the difference between, okay, there’s Xreal glasses powered by Bose with a Bose logo on the side, and selling the technology without the Bose logo? Or does a Bose logo always come with it?

When you said this is new for Bose, in a way, yes, but in a way it’s exactly what we’ve been doing in automotive for over 40 years. And so we do have a model for doing this inside the company, and it’s just broadening that thinking into new spaces. First and foremost, we’re selling technology, and the technology is the key. And then in some cases, we’re also selling the brand. So you might get technology in a product without a Sound by Bose or Powered by Bose logo on it, or you might get the logo as well, but you’ll never get the logo without the tech, if that makes sense.

Some of our partners are interested in the brand, some are not. Sometimes we’re interested in having our brand connected to their brand, and sometimes we’re not. And so that’s part of the discussion we have with each of our customers, about whether it makes sense for them and for us. And so when you see it, know that it’s been really thoughtful, but it’s also attached to a lot of great technology that’s inside that product.

Tell me about those conversations. Is it cheaper to take the brand? Is it more expensive to take the brand? This is one of those places where Bose has maybe a bigger, more famous brand than maybe some of the customers. How are you thinking about that, and how does the cost change?

As you would expect, the technology has a licensing fee to it, and so does the brand. We’ve invested over 60 years in our brand. We’re incredibly proud of it. It is, I think, one of the most iconic brands in the world, full stop, and certainly within the audio space. And so we are incredibly careful about where we put the brand, and it certainly isn’t free when you put it on a product.

Right next to that is McIntosh. As you know, I grew up a total stereo nerd, just reading the magazines, dreaming about having the blue meters in my house someday. Still working on it. I’m going to get there. I did have a Jeep with a McIntosh sound system several years ago. The blue meters were… It was a very fake Android app that ran on a very slow Android infotainment system, but I still loved it.

That’s a brand that has a real halo. You can’t make cheaper McIntosh products. That’s not good for that brand.

No way.

You need to stay way up there. How are you thinking about that relationship to Bose, which is much more of a premium mass market brand?

We think of them as two completely separate brands. And then think about Sonus faber as yet another third brand, all of which are targeting really passionate fans of those brands for what they stand for. So when we thought about getting into the luxury space and acquiring those two brands, we spent a lot of time saying, “What is it that we’re going to do internally to protect the brands?”

We do not want to Bose-ify anything that is happening in those brands. And so we’ve set it up as a separate business unit. We have a luxury audio business unit that houses those two brands. We are very careful. As you said, we have no intention of bringing products down-market. We believe both markets are exciting. We’ve been in that premium audio space for a long time.

When we looked at what we’re calling luxury audio, we had two really important insights, I think. One is that as we’re doing our own technology work, there’s a lot of really exciting technology we develop that just isn’t right for a premium audio product. It’s right for something that sits at a higher level. We just didn’t have anywhere to put it. And that was true on the consumer product side, and it’s certainly true on the automotive side where we felt like we were bumping up against what you could put in a premium product, and the ceiling of what we wanted to do was much higher.

So the first reason we’re excited about it is that it sort of unleashes the creativity of our engineers to think about what you could do at those higher price points with just a different type of customer, a different type of experience. So we were excited on the technology side.

We were equally excited, I think, about this idea of luxury. If you look at the luxury market writ large across all categories, what you find is that audio is really underrepresented compared to the role luxury plays in automotive, hospitality and travel, fashion, or leather goods. And when we really sat back and thought about it, given how much music is a passion point for people, it didn’t make sense to us.

We think there’s a huge opportunity to attract new customers to the category if the experience feels more like a luxury experience. And so when you think about those products, it’s certainly about how they sound, but you went right to it. It’s the iconic design, it’s the artistry, it’s the craftsmanship. All of our luxury products, whether it’s McIntosh or Sonus faber, are handmade. They are crafted one by one for you as the customer. And there’s something incredibly special about that.

These weren’t your decisions, so I feel like I can be a little bit rude about them. When I saw McIntosh was licensing to Stellantis to be in Jeeps, I thought, “This company has to be desperate for money. That’s not actually a match.” It was fun that I had a Jeep with blue meters in it. It was great.

But I thought to myself, “This is a total mismatch, and they should be in Maybachs,” or whatever it is, whatever luxury sector that you want to be in, because that’s the brand. Was this just opportunistic for you that this was a company that needed the help, that was in trouble, that you could go and you could push your products upmarket without having to take the Bose brand there? Or was there some other opportunity that presented itself to you?

We weren’t actually shopping for brands first. The first thing was, strategically, we saw the opportunity in the luxury space. We saw an opportunity to create something new around luxury audio, and we were really excited about that.

So once we had made the strategic decision to go there, we had a lot of conversations about buying a brand or creating a brand. But the reality is, in luxury, at those price points, history and heritage are part of what you’re buying. And so we sort of intuitively knew we wanted to find a brand or brands that really had that strong history that we felt like we could build on top of.

We thought those were both brands that had incredibly passionate fans, and we saw an opportunity for us to take them to a new level. So that was how we ended up acquiring the McIntosh group.

Bringing in their engineering organization, they have a lot of technology. How integrated has that become over time, or are you leaving them alone?

We’ve brought it into our engineering organization, but we’ve also kept a lot of the aspects a little bit separate. So probably the best example of that is Bose has a Bose sound. Everybody knows that. You buy a Bose product… You buy one, the headphones I’m wearing, the speaker in your house, your car. There is a sound signature to brands that consumers expect. That comes from the engineers.

The engineers are tuning products to sound like they should. And so we’re very careful on the engineering side to protect the sound signature of McIntosh. What we don’t want is for you to hear a McIntosh product or a Sonus faber product and imagine that it sounds different. So that’s one.

Initially — as you do with, I think, any acquisition — you want to be careful to protect what you bought. We’ve integrated it, yes, but we have a separate team that is doing the engineering for luxury, and we’re slowly adding Bose technology to it where it makes sense. So there are things, for example, on the Bose side where we can be incredibly helpful in terms of the software aspects, the connectivity, those sorts of things. Those are things that we do really well, and they’re not the most important part of those products, but they have to work. There are aspects that we’ve started to add into the products, and you’ll see more of that over time.

I’m really curious about that luxury aspect. The knock on high-end audio for years, and I’m not saying this is McIntosh, but the knock on high-end audio for years is that it’s the same stuff, all dressed up fancy. There are headphones that are famous for having weights in them to feel heavier even though there are commodity parts inside.

This is one of those places where you could just say, “We’re just going to make more expensive versions of the same thing.” And some buyers are totally price-insensitive, in that they will always just buy the most expensive thing. Then there’s, “We’re going to make it worth it. We’re going to make it worth 10X the price of whatever Bluetooth speakers are at Best Buy.”

Where are you falling on that? Because it’s very tempting, especially in luxury, to just make it expensive.

The value comes from the sound. You should come visit us at Bose in Framingham and listen to a McIntosh system, a Sonus faber system. They’re incredible. They’re not anything like what you would get at a Best Buy in a premium product from Bose or anyone else. And that’s incredibly special.

Part of that is the technology itself. Part of that is that it’s more analog, so it has a different sound signature to it. It’s got a lot more power behind it because that’s what McIntosh is known for, and that’s what you need to power a Sonus faber set of speakers. You’re also paying for design. In a way, it’s a little bit of art.

We just launched a new integrated amplifier, the MA2375, for McIntosh. I’m so excited about it. It’s sort of got that retro look with the vacuum tubes. It is a design piece. I want to sit it on my desk. I want it to be a part of the decor. You would say the same thing on the Sonus faber side. It’s like buying an incredible piece of Italian furniture that has been handmade with natural materials and just feels like art that also sounds beautiful.

So there’s more to the value proposition in the luxury space. It’s true in the premium space too. Design really matters, color matters, feel, touch, comfort, all those things matter in addition to the sound. I think it’s just even that much more elevated in the luxury space.

I’m really curious about all the different sets of new customers you’re describing. There’s the B2B business, which is growing in audio tech. Then there are luxury customers who have a lot of money, but they’re fickle. And particularly in high-end AV — and I’m a high-end AV nerd through and through — convincing people it’s worth it because they can hear it has always been the hardest problem.

Maybe the most fun problem, because you can convince a lot of people they’re hearing a lot of things that aren’t there. You can sell expensive cables at a hundred times the price of the regular cables, even if they don’t do anything different. This is a lot of new customers for Bose. Do you think the B2B business is going to distract your core technology roadmap? How are you holding that all together?

It is more complex for sure. It is a broader mission, and I think we’re ready for it. When I was here last time, we were in the middle of the transformation that I would say was solidifying the foundation of Bose, and now we’re building on top of that. We’ve built an amazing marketing and brand infrastructure that we’re now using to power three brands instead of one, but a lot of that capability can be put in the direction of all three.

So the key is real clarity about who the customer is, which we’ve done a lot of work on the Bose brand for sure. We talk about the music lover for Bose, but we’ve done the same work for McIntosh and the same work for Sonus faber so that we are one thousand percent clear about who the customer is and how we want to talk to that customer.

On the kind of products side of our company, I think we’re really clear about what the mission is for each one of those brands. On the technology side, interestingly, as we become more of a B2B technology company, it’s actually creating advantages back to the Bose brand and our luxury business as well. And what I mean by that is that it’s caused us to completely rethink the way that we do research and we develop our tech.

In the past, if you think about Bose developing products, we were developing purpose-built technology for a pair of headphones or a car audio system. And I think last time we talked about the tech franchises. We care a lot about noise cancellation. We care about immersive, lifelike audio, and we care about this idea of hearing what you want. And those are all experiences that are built on some basic building blocks of technology that we’ve been developing for a long time.

But because we were developing them for our own products, we’ve had to do a complete shift in the way that we think about it so that we’re now developing, from the ground up, those building blocks. Now we’re able to say, with common platforms, common algorithms, AI — think of them as sort of AI units — we can create different experiences for our own products, and we can port that same technology through the same platform tech into other people’s products.

What that’s creating is a lot of efficiency at the research and development level. It’s creating more breadth because now that we’re thinking about things like noise cancellation for different spaces, spatial audio, or hearing what you want, it is actually creating more innovation and more invention at the core. It’s making us faster because we’ve made these building blocks a lot more consumable for our B2B customers; it makes us faster at bringing our own products to market, too.

I think this is a great place to ask the Decoder questions because you’re describing the endpoint of a pretty big culture and structure shift at Bose that you’ve been architecting for several years. So how is Bose structured today?

There’s a business unit structure, there’s a functional structure, and we are living in what I would describe as a hybrid. So for the core functions that matter deeply to what we deliver, like engineering, marketing, and supply chain, those are global functions that support all of our businesses. But because we’ve added new customer sets and really two new businesses with luxury and audio tech, we’re also running four business units. So those four business units have business unit presidents, and they draw on the global functions for the work that they need.

How many people is Bose today?

About 2,500.

And you’re mostly in Framingham, or are you all over the place?

We’re global. All over. Yep.

You said you’re a hybrid. How’s that split between engineering, sales, all that sort of thing?

I don’t actually think about it that way, but the one thing that I do know — sort of the most important statistic to me — is that half of our employees are engineers.

You’re describing AWS, you’re describing Amazon. You’ve modularized a bunch of components inside the company; now you can build new products faster. This is really great. Every company that goes through this transition ends up hiring more and more engineers over time because you just need to build more and more components. Has that been growing? Are you investing more in that kind of engineering?

We are for sure. I think it’s probably not unique to us. We’ve been in this transition of hardware long, long ago. It was just hardware, and then hardware and software, and then more and more software. And now as we’re building the technology pieces, those are truly just software products that are an algorithm- or AI-based product that we’re selling with.

You see that mixed shift continuing. We’re certainly seeing growth in our engineering ranks. And we’re getting a lot more productivity out of engineering as we start to have more sophisticated tools, as we use AI and other things to help with some of that development. So yes, growing. And I think more productivity per engineer, if you will, as we start to use more sophisticated tools to do the development.

You said it’s pure software products, right? When you sell out to B2B, you’re just selling the algorithms. Do those feel protectable? This is one of the things that I think about all the time. What is the actual thing you’re selling, and how easily can it be copied by a Chinese firm that does not respect our intellectual property laws or something like that?

Yeah. I’m probably oversimplifying. We’re really selling three things. We’re definitely selling software, algorithms, et cetera. I’ll come back to your “how do you protect it?” question. One of the very special things about Bose is actually systems engineering, which is an underappreciated discipline in my view. So software’s great, but software has to run on hardware in our world. Ultimately, that software has to get delivered through something.

The magic at Bose, the secret sauce, has always been systems engineering and how you’re marrying together hardware and software elements for an experience that feels unified across the device and that’s optimized to provide the best experiences. So we’re certainly doing that on our own products, but you see that kind of a service as we start to sell more and more technology into our B2B customers.

There’s also a lot of desire for help on that system side because our expertise there is real. That’s always been true in our automotive business, and it is increasingly true in our partnerships and our customers on the audio tech side as well. So I think that the second piece, which is really important to call out, is that it’s not just software for software’s sake. It’s software that has to be integrated into a system where everything comes together to work. And then you’ve got kind of the idea of systems engineering and then services on top of that.

So, how you tune a product. Some of our partners and customers need help with that, want help with that; some don’t. There are a lot of services that get married along. Sometimes it’s systems engineering. Sometimes it’s things like tuning, et cetera. So three pieces.

Can I play devil’s advocate here for one second? What you’re describing here, if you wanted to poke at it as hard as you could, is a process of a little bit of commodification. There are a lot of common components here, and then it’s the layers on top that differentiate in the market.

If you’re an audio nerd, the magazines are all saying one engineer in their garage innovated on speaker technology, and they made the speaker cones out of waffle paper, and now they’re more transparent than ever. Whatever that myth was that sold a lot of speakers back in the day.

If we’re not starting from the products back, but you’re starting from the components up, there’s a risk that everything kind of becomes the same, and it’s branding and marketing, and maybe the sound signature that’s tuned into it is artificial now — it’s not from the components themselves. How do you balance that? Or is that just not a worry with modern consumers?

I think what you’re describing is that if we’re putting Bose technology in other headphones, doesn’t that just revert to the mean? I think there’s a second, bigger market for us that is not audio products. Things that we call audio second. So for example, the XR glasses, the main function is not audio, but audio is really important. You don’t usually buy a car because it has a Bose sound system, but when you buy the car, the sound system is actually one of the most important features of the car.

A projector that you buy for your house, you’re buying because it’s a great projector. We’re partnering with Epson, and the sound is also really important there. So it’s a little bit different, I think, when you get into audio second. In the audio market, I think that the key is that there are so many different customers who want different things.

When you think about products that have ingredient technology that’s the same, it’s really enabling you to do things for your customer and the experience that you’re trying to create. I’ll just use Skullcandy as an example. Skullcandy just launched a pair of headphones with Sound by Bose, and you couldn’t have two more diverse customer sets between who they’re targeting and who we’re targeting. We helped them with the technology. So there’s Bose technology inside to help make that sound better, but the design choices of what they want for their customer are incredibly different from what you would find [with Bose].

So I don’t think you would ever put those two products, a Bose headphone and a Skullcandy headphone, side by side and say, “Wow, this is really commoditized. It sounds the same. It is the same.” They’re very, very, very different.

I think I’m asking slightly more about the audio-second product. If all of your audio-first products come from the same technology base, the same modular components that can be remixed into all the products you want, do you lose the ability to innovate at the product level? If everything is made out of the same Legos, at some point you can’t be like, “Actually what I need is all new Legos.”

I think the answer is no, because the product is so much bigger than just those ingredients. So if you think about headphone design, the comfort and all of the other features, right? There may be a lot of ingredient technology, but the ones that you choose may be different.

One of the things that we talk a lot about now is being at the intelligent edge. Everything Bose does is what we would describe as at the intelligent edge. And as you know, it’s a pretty hostile environment. You have limited compute, you have limited power. You’re always making trade-offs between what I am putting in this product and how does that impact my ability to do that? Because there’s just not enough room to put everything in it.

So I think the reason why you still see so much differentiation among products, even if some of the ingredients are the same, is because all of those product choices that you’re making are really critical, and you have to make a lot of them to fit inside that constrained environment that you’re working with.

This brings me to the other *Decoder* question about decisions. You’ve obviously made a lot of them, and you’ve got a few big ones to make that I want to come to concerning AI. How do you make decisions? What’s your framework?**

I am a believer that you get the right people in the room, you have an intense debate that isn’t always comfortable, that sometimes makes people a little bit queasy, and you move as fast as you can to decide, knowing that if it’s wrong, you can pivot. That is, in my simplest terms, the way I think about it.

We get a lot of type one and type two decisions, move fast, in that we had an ex-Amazon employee write to us and say, “Every decision is reversible if you have enough money.”

[Laughs] That might be true for Amazon.

That’s a very Amazon perspective. The reason I bring this up in the context of AI is that I’m looking at the sweep of things you’re doing, and I’m looking at where, in particular, the big tech companies think their headphones are going to go.

I was reviewing our conversation from four years ago. Boy, did we talk about Bluetooth a lot. We did. This is a core limiting technology. It’s an enabling technology in an important way, but they hold onto what Bluetooth can and can’t do on their platforms very tightly. And then their products, their first-party products, get to extend that standard in different ways — and they all do it.

This broadly means if you have an iPhone, the best working headphones that will do all the things the platform wants you to do are probably AirPods. If you have a Pixel, it’s probably Pixel Buds. If you have a Samsung phone, it’s probably their buds.

And this is because they have reduced the capability of the interconnect. One of the reasons they’re doing that is because they want your earbuds in your ears 24/7 so you can talk to their AI systems on the back end. And this is their future. Maybe the answer is that we’re all going to wear Meta Glasses, or we’re all going to wear Xreal glasses. It seems up for grabs in big particular ways. What’s your view of that? What’s your view of AI wearables? And do you think there’s room for third-party players here?

We absolutely think wearables are critical to the AI future. I think that we’re clear on that. The form factor, if you work backwards from that, we’re going to use our wearable devices to communicate with our AI agents. And that is going to become more and more important and more and more natural. We strongly believe that.

The second thing we believe is that it is more likely to be conversational and sound-based than anything else. At least we think that is probably the primary mode. And so the types of technologies we’re focused on are around voice pickup, clarity, speech enablement, and we think all are incredibly important. What form factor you want to use as your wearable, I think the jury is still out. I think the answer is going to be a whole proliferation of them. Some people will like glasses because they like the glasses.

Some people will like open earbuds, which we hadn’t launched when you and I talked last time. We talked a lot about open ear audio, but not yet about that form factor, that clip-on form factor that Bose invented and launched into the world. We think that is a really powerful form factor that will be for some. Others might want a watch, a pin, a ring, or 18 variations of how those work together.

Part of the reason we think being in the audio tech business is important is that it gives us the ability to participate in a lot of those different places that we might not naturally be able to go on our own. So we think form factor is going to matter, but I don’t think there’s a single winner there. At least that’s my current view. I think that there will be multiple types of devices that consumers will choose based on their own personal preferences.

We know there are going to be a limited number of LLMs, right? We know that those are going to converge to the large ecosystems that already exist. New ones may emerge, as we’re seeing, but there won’t be hundreds of those. There will be few of those. And yes, there may be some advantages if you stay within that ecosystem for your earbuds. But fundamentally, I think even though a lot of people choose earbuds, headphones, and other devices in their ecosystem, a lot of people don’t.

I think consumers still want choice. And I think it is important for those ecosystems. In our conversations with them, they need third-party players to interface with their AI as well. They need third-party players to produce devices that work with what they’re doing because they recognize that they’re not going to have a full share of the market and that choice matters.

So we believe we’ve got a couple of ways to compete now. Our own devices, for sure, need to be compatible across the different LLMs and the ability to interface with them. And we need our technology starting to crop up in other places where that interface will happen.

Is this a way to just hedge the bet that no one actually knows what’s going to happen with AI? If you’re selling Bose beamforming technology to a handful of Silicon Valley startups making pendants, that means you’re not making the big bet on pendants being a form factor yourself. You can wait and see. Is this just a way to make sure the technology is available and being monetized while you wait for the answers to emerge?

There’s certainly a diversification play, absolutely. And I wouldn’t shy away from betting on a few, and if one becomes a winner, awesome. I also think that it’s important not to lose sight of what your brand is. The Bose brand is: we are for music lovers. And if we start to branch out and create some of those other devices that are not core to who we are, I just don’t think it works. So it is a way to participate because we don’t think those are areas that it makes sense for the Bose brand to go.

One of the things I worry about, again, as a music fan, as an audio nerd, is every single time the tech companies come to music, convenience wins, and quality suffers. This is the pattern that has been repeating since MP3s first came out. We’re going to go from CD quality audio, which in 2026 is still what people refer to as the best audio you can get. CD quality — that’s the bar.

But we went from that to, I don’t know, 128 kilobit MP3s on iPods. We stayed there for a long time. We got a little bit better. The headphones got worse. The interconnect went from wired to a compressed Bluetooth interconnect. At every point, the market picks convenience. There’s some point here, even in luxury, where you just walk in, and the music is playing because the AI has determined the music should be playing, and the endpoint of the music is the least important thing. It’s just that it’s happening to you.

There’s not this focus on listening intently because you’re probably also watching TikTok at the same time. Or maybe you’re only listening to music on TikTok, and that’s where all your discovery happens, and all of your music money is spent at concerts and live experiences. Somewhere in there is the audio industry. It’s been going through this convulsion.

There’s another turn coming here where Apple, OpenAI, Samsung, and whoever else are going to launch AI-first devices that are conversational, that have audio input and output capabilities. And they will say, “And you can also listen to music.” They will commodify it even further, and the idea of a music lover gets even more squeezed. How do you view that? Because as a music lover myself, I’m like, “This is only getting worse. It’s only becoming. It’s not audio second. It’s like audio 15th.”

Okay. So it’s a complicated web there that you just spun.

Yeah. I’m having an existential crisis here.

I’m going to try to help you. In some ways, what we find with our luxury customers is that what they’re looking for are moments of clarity, quiet moments where you put down the phone, you put on a record, right? You listen to it the way that it was supposed to be heard on an amazing sound system. And the quiet nature of that, the ability to step away from what is a chaotic digital world — that trend is picking up. It’s picking up. We see it in the luxury space. We actually see more of it in the premium space a little bit now too.

I do think that the appreciation of music, just appreciating it for the emotional state that it puts you in and what it does for your mental health. I actually think that is a trend we’re seeing. And I actually think that’s going to continue to grow.

Our belief and part of the reason that we started this audio tech business is that we want you to have the best sound experience no matter what. Even if you’re listening in a hostile environment, you’re listening on a lower-cost pair of headphones, or you’re listening to a crappier format, let’s at least make sure you have the best experience possible and that you can tell the difference between what good and better sounds like, or what better and best sounds like. So I think that’s the second thing.

The other thing that I think is positive about what you said is there’s just so much music discovery because it’s everywhere. And not everyone is going to turn into a music lover, and that’s okay. But I think for young people who are discovering, I’m always amazed at the songs my kids are excited about because I’m like, “You know that’s not a new song, right? That’s from when I was a teenager.” I think that’s great.

The way that the younger generation is doing music discovery is creating a love of music that is not bound by time or genre. I think that’s really cool. And our belief is that if you help people listen to that as best you can, they will get an appreciation for what better sound sounds like, and over time they will graduate into the formats and the products that actually bring that to life in an amazing way.

But that’s music. I appreciate that this is happening in music. I think the question I’m asking is about AI and the almost religious belief in Silicon Valley that AI, in particular natural language input, will usher in a platform shift. Mark Zuckerberg has a religious belief that you’re going to talk to Muse Spark through Meta Glasses.

Whether or not you believe him, he is laser-focused on that because he wants to stop being constrained by Apple’s platform policies. That is the story that plays out in the pages of Verge every single day.

Apple probably believes this, right? Apple believes natural language input is some kind of enormous threat to the iPhone. They’re reacting to it. Somewhere in there is, are you wearing headphones to talk to this? And can Bose headphones meaningfully participate in those ecosystems? Or will all of these companies create more first-party opportunities for themselves?

I think both can be true. So I think we’ll see if Mark is right, but that doesn’t mean you want to have a crappy listening experience when you’re listening to music. Do you take the glasses off and put on headphones? Do you make the glasses sound better? I don’t think that just because we have a utility, which is AI on the… I’ll just use phone calls. It’s an old school example. Just because you can take phone calls on your headphones, if they didn’t sound good to play your music, most people who love music would not wear those to listen to music. It’s just the reality. And I think the same is true. It’s a utility.

If you’re someone who doesn’t care so much about music and it’s not part of your personal brand or your identity, you may say, “That’s good enough for me. I can’t tell the difference.” But I think for most people, they will choose multiple devices or they will choose one device that can do all of it really, really well. And I still think there’s room for high-quality listening experiences, even in a world where we’re all chained to our AI.

So I’ll just paint a picture for you. If Xreal wins the AI race and all of us interact with Claude all day long through our Xreal glasses, and the sound is powered by Bose through an audio tech partnership, is that victory? Is that success? Or is there more to be done?

I think there’s always more to be done. When you think about how fast technology is moving, I think we’re imagining experiences today that can be had at the edge, that live within the compute constraints and the battery constraints that we have. Two years from now, those constraints are going to still be there, but the amount of capability you have will be larger. I think we’re going to continue to see innovation and invention in the space that’s going to create experiences we haven’t imagined yet.

One of the things that I’m really curious about is just the amount of audio input all of these companies want to do. We had the CEO of Zoom on Decoder, and he was like, “Everyone will talk to AI avatars in Zoom and all of us will be mediated by Zoom. And that means we’ve got to record the in-person meeting. Everyone has to wear the rings or the pendants, and we’re going to record everything all day long so the AI can fully know us.”

I have a lot of thoughts about that, but that’s one vision in the world. We’re going to record everything, which sort of puts a lot of pressure on noise cancellation, on rehabilitating the audio once you have it. There’s not just a lot of noise, but damage that happens in the compression process and transmitting over the internet. That all seems like an opportunity for a company like Bose.

Is that where the investment is to help the AI hardware companies deal with lossy streaming and TikTok audio? What do you think about that opportunity?

One of the things that we’re learning as we further venture into this space is that the core problems are the same. If you take something like speech clarity, voice pickup, that is a world we’ve done a lot of research on, a lot of work on. It’s an incredibly important part of a Bose headphone, for example, but it’s also important in a dozen other places where the same type of technology, at its core, can be expanded and extended.

So think about driving 70 miles an hour down the highway in my car. That same speech clarity technology is really relevant in the car in the same way it is in a headphone, in the same way that it is in a motorcycle helmet, where you can imagine that’s maybe the most hostile sound environment you could get.

I think the same is true. Noise-canceling has all sorts of varieties that, I think, are incredibly important in the AI examples that you talked about and in a lot of other everyday life examples where noise cancellation hasn’t gone yet. A great example of that is hearing aids. As you know, we’ve been working on hearing aid technology at Bose for over a decade. Just this year, we launched the world’s first hearing aid with a customer and partner called ORKA in China that has noise cancellation.

It actually now solves what I think is the biggest problem that hearing aids have had for decades, which is that my most challenging environment, if I have some hearing loss, is a noisy restaurant, for example. And traditional hearing aids amplify all of that sound. Noise canceling has always been the dream of hearing aids because if you could take the noise of the environment down and amplify the speaker, the person you’re trying to have a conversation with, that’s sort of the holy grail now because you’re doing exactly what the customer really wants and needs.

That, up until now, has been impossible to do because in a hearing aid, you need 15 to 18 hours of battery life. You have very limited compute. It’s been a really tricky problem. For us, solving that problem was both about hardware — a new driver that actually fits in the ear canal, but can power noise canceling — and the platform that we built to enable that at a low enough compute level that you could do it inside of a hearing aid over a period of 15 to 18 hours in a day. That’s the kind of innovation for noise canceling, for example, where it’s an obvious use case, but to get there required a considerable number of building blocks.

I’m glad you brought up hearing aids. The last time you were on the show, you had a partnership, I believe, with Lexie. And after our conversation, I ran out and bought my dad those hearing aids. After four years, he won’t wear them because he says everything is too loud. Exactly the problem you’re describing.

But you exited that partnership with Lexie, and the parent company — just asked a couple of weeks ago — is leaving the US market. They’re out. You’re describing hearing aids that you’re making in China. Is there something particular about the US market that just makes it too hard, or are you going to reenter?

No, we’ve been here with Lexie, and they’re going through some challenges, and so we’ll find the right partner for us in the US. I think the over-the-counter hearing aid market has struggled more than any of us thought. If you go back four years, it hasn’t taken off at the rate we had hoped that it would. But I still see a ton of opportunity in the hearing aid space. And this hearing aid happens to be in China, but there’s no reason that you won’t see it in other places around the world.

Why do you think it hasn’t taken off?

I think it hasn’t worked yet. There are a couple of factors. And again, we’ve been partnering into it, so we’re not as close to the day-to-day. But I think there are really two things. It’s definitely cheaper and more accessible than it was in a prescription environment, but these are still very expensive products. So, rounding about a thousand dollars, which is a lot of money.

I think in an over-the-counter space, most customers are expecting something that looks and feels more like what they would be willing to pay for a consumer electronics product, which is more like two, three, or maybe $400. While it’s more affordable, it is not as affordable as I think the consumer wants, needs, or expects. I still think there’s some work to do there from a technology standpoint to figure out how you bring it down.

It’s also not really in the insurance space yet. If I get a prescription hearing aid that’s covered by Medicare or that’s covered by my insurance company, but that’s not the case for over-the-counter. And so I think there’s just some affordability challenges we haven’t quite fully solved yet — ones that the industry has not solved. That’s one piece of it. I also think there’s still a stigma part of it. There’s just still a, I don’t want to look old. I don’t want to feel old. I don’t want to wear a hearing aid. And until we figure out the right industrial design and make it more palatable, I think you still have resistance no matter the price point.

The hearing aids my dad was most resistant to were when I tried to get him to wear AirPods, after Apple had released that feature. He looked at me and said, “Absolutely not.”

So this is a particular area of convergence. What is an augmented reality hearing device except a hearing aid? Except a hearing aid with intense noise cancellation that can focus on the speaker in front of you and maybe translate them in real time if you don’t speak the same language? You’re talking to an AI, and you just want to whisper so no one can hear you, and it’s talking to you, but you’re still hearing everything that goes on.

This is the convergence, right? The idea of headphones, the idea of augmented reality, and the idea of hearing aids, they’re in a Venn diagram. And the center of that Venn diagram, I think, is getting bigger and bigger over time.

I just keep coming back to the same question: Is the culture of Bose designed to provide technology to all of the players in that ecosystem? Or are you still trying to win with the best product that hits the middle of that Venn diagram?

We are always focused on who the Bose customer is and what the features they want are. As I’ve said a couple of times, you can’t fit everything in. That Venn diagram is getting bigger as compute is getting more powerful at the edge, but it’s still limited.

When it comes to adding AI, we talk about it at Bose as tiny AI. A lot of companies are good at really big AI. What we’re good at is taking really complex algorithms and then figuring out how to shrink them down and take as little power and compute as they possibly can so that they can fit in our devices. But you’re always constrained. So we’re looking at what our customers are asking us for, what they need, what they want in that Venn diagram. Where do they want the emphasis to be? And that’s where we’re pushing our products. You’ll continue to see us do that.

Other companies will choose a slightly different variation of that Venn diagram, and they’ll look for a different feature set. And I think that’s what makes the industry go. It is competitive by nature, and it is important to have a deep understanding of what the customer needs are.

Do you think that there’s any pressure on Bose just from the product manufacturing side? There are tariffs, there’s a lot of focus on what gets made in China. We talked briefly about protectability and whether or not you even want to sell into Chinese companies or Chinese OEMs who might turn around and take your technology, which we have seen at scale with Apple, Tesla, and others. Is it worth it to be chasing hardware at this time?

I think, since COVID, the supply chain has become a more exciting environment.

[Laughs] Sure.

So it’s been a series of things. I think our perspective back then was that we said, “Look, it’s clear we’re going to need to be more flexible and more agile in our supply chain.” That, I think, became clear. We’ve done a lot of work to get ourselves into a position where we expect curveballs now; we just don’t know what they’re going to be. And when they come, we’ve designed ourselves in a way that we can adjust quickly, and we think we’re flexible enough to handle it.

The world needs hardware. As much as everyone is enamored with software and AI, and those are the things creating the magic, perhaps, on the hardware these days, you still need physical things. We are a physical world. You need physical things, and we are going to continue to make physical things and roll with the curveballs that come our way.

Have you moved manufacturing, meaningfully, out of China?

We have, yeah. So we’re diversified across Asia. We also have manufacturing in North America for some of our products as well.

Have you moved anything, meaningfully, back into the United States?

The only thing we have in the US is McIntosh. So all of our McIntosh products are built in Binghamton, New York.

It’s funny, because those are the only products you have that also wouldn’t benefit from the price cut of not having tariffs. You need to keep those prices really high. Have you thought about moving anything back to the United States to avoid whatever tariffs and bring prices down?

We’ll always look at it. It’s a fulsome economic equation of not just economics, but also expertise. Consumer electronics expertise from a manufacturing perspective moved to Asia a long time ago. And so it’s both a cost and an expertise question. We look at it, we talk about it. If it ever makes sense, we’ll do it. Right now, those things don’t play out for us.

I feel like I have to ask you about Bluetooth one more time here at the end of our conversation. I’ve waited.

It is the limiting factor of so many things you are doing. It’s the limiting factor of how you can connect to phones. It will almost certainly be the limiting factor of how all these AI devices work.

The manufacturers have really put Bluetooth in a box. There is a wall around what it can do and what they will ever allow it to do on their platforms. And then there are their extensions. And this is across the board. I’m not even singling out one manufacturer. I used to always single out Apple, but now they’re all doing it. So it’s like, great, there’s a wall around Bluetooth.

For a company like Bose, which has to connect to phones, that is the limit. Have you tried to push hard to overcome that limit? Have the antics of the European Union helped you open up and overcome the limit? How has this played out over the last four years?

Probably not terribly different from what we talked about last time. We partner, we push, we ask for what we want, and we work with what we have. So I think we are always trying to test and push the limits as much as we can, and we’re incredibly good at not using that as an excuse. So how do we create amazing experiences with the advantages that we have? And we think we’re pretty great at doing that.

On the other side of this, you’ve got McIntosh products, your premium home products, and people are disconnected. A lot of people still want to play Spotify off their phone onto their home stereo system. There’s a joke we tell in our other show, the Vergecast, where we just try to locate the computer. Where is the computer supposed to be here?

If you’re playing music off your phone, is the computer your phone, and you’re just Bluetooth streaming to your speakers? In the context of McIntosh, that is just like a heartbreaking crime. Don’t stream Bluetooth to your McIntosh system. So that means the McIntosh system needs a computer. It needs to run some variant of a Spotify app that has Wi-Fi that pulls lossless down. That means it needs a ramp. The whole thing immediately becomes a computer. Maybe a somewhat low-powered computer that’s only focused on a handful of apps, but a computer.

You also have all the carrying costs of a computer. Software updates, application updates, Spotify changes its file formats. Whatever is going to happen, you’ve got some problem there. How do you think about that?

Across all of your products, the amount of compute going into them is going up, as you’ve described. Maybe in tiny ways, but it’s going up. And then you’re not charging service fees. How does that all work as more and more of your products fundamentally become network computers?

I think consumers are subscription’d out. I think that the idea that you’re going to create a subscription service over an over-the-air update that they’re going to pay for on a monthly basis, I think that’s not coming in our space. So we have to kind of think long term as we’re selling products today: what is the software, compute, cloud storage, all of the security, all of the costs that live across the life cycle of that product that we’re going to have to support. We have to build that into the cost of maintaining that product, right? Which is different from what you would have done in the past.

We’ve made some design choices, I think, that are important as well. We don’t need to duplicate what Spotify is doing either. So as we think about the lifestyle collection products that we just launched, what we heard overwhelmingly from our customers was, “I just want to go to my music. I don’t understand why I have to deal with you. I love how this sounds, but please just get me to my music faster.” We thought for a long time about how best to do that in the home. And we think about it in our wearables too. We want you to be able to control your home speakers through Spotify, AirPlay, or however you want to do that. And that offloads some of that from Bose, which is good.

We also just made an acquisition of a company called Stream Unlimited. Their business is to connect to all of the streaming apps, devices, et cetera, that allow a smart home device to have those easy connections. That gives us some scalability and some participation in that orchestration layer, which is also helpful for us. So we come at the problem in a couple of different ways.

There are rumors, in a few months or weeks here, that Apple will put out a whole bunch of new smart home devices. OpenAI might put out a smart home device. Obviously, I have Google ones, and there are Amazon ones. You ask any of those companies, what are the number one and two things people ask for from their assistants in their house? And it’s setting timers and playing music. Then everything else is a long tail of almost nothing. The drop-off is enormous.

If you want to play music and you’ve got a great stereo in your house, you’ve got to go connect to their smart home system. Now you’re on their network, and now you’re just another endpoint for whatever forthcoming successful consumer AI agent will exist. Big question mark there, but let’s say it happens.

That further commodifies you, right? That’s yet another time when they’re not going to use your interface or your streaming; they’re just going to say [a command] out loud in their home, and the OpenAI device will be like, “I found some speakers on your network. I’m playing the music for you.” Do you need to participate there? Do you need the OpenAI agent to say, “I’m playing it on your Bose speakers,” and then you collect a tiny little licensing check back, or how should that work?

I don’t think so. I think we understand there are a lot of components in this chain, and customers think about Bose for amazing sound. And when they’re playing the speaker, and they’re asking whatever agent they’re asking to play it, there is an appreciation. There is going to be a logo on that speaker. If you have a new Bose Lifestyle speaker, there’s going to be a logo right on the front of it. You’re going to look at it every day in your kitchen, living room, or wherever it’s placed.

I get the feeling those logos are about to get bigger and start lighting up. You can see it coming.

[Laughs] We’re not going to make them that big.

So that chain is long. And in fact, you launched another piece of the chain of music, which is that you launched Studios, which feels like a record label, a marketing venture for new artists. How does that play into it?

So bigger than a record label, not just a record label, but that’s a piece of it. Think about Bose Studios as a way to connect with our customers and give them new avenues to participate with us. The announcement may feel like a sudden thing, but if you sort of look at what we’ve been doing at Bose over the last four or five years, we’ve been moving in this direction.

Getting the attention of consumers is hard. You don’t do that with traditional advertising anymore. Consumers are engaging with content, and that’s it. And so we’ve been building more and more content. We’ve been partnering with more and more creators. We’ve been dipping our toe in this water for the last several years.

So I’ll give you a couple of examples. We’ve been partnering with NME to do their mixtape, where we’re helping them bring undiscovered, up-and-coming artists to the forefront and helping bring and celebrate those artists. We’ve learned a lot from doing that. The connection and the participation of our customers with that content and with those artists has been incredible.

We’ve been sponsoring an event, the Chris Brickley Invitational, which is this amazing basketball event for high school basketball players. It is becoming the place to go if you’re an All-American type high school basketball player, both on the men’s and the women’s side. And that event has turned into something so special. We’re not putting on a new Super Bowl over here, but we are participating in content creation in a way that connects with our customers. We think that’s really important.

We also like this idea of up-and-coming creators who need support to launch their careers. At Bose, we say fiercely independent. There’s something so unique about our structure and the independence that we have. We kind of like the idea of supporting this up-and-coming group. It supports our connection with our customers, and we think we’re helping them along the way as well.

I feel like I could spend another full hour just talking about the economics of the music business and brands supporting creators in this way. But sadly, you’ve already given us too much time. Lila, thank you so much for being on Decoder.

Thank you for having me back.

• Questions or comments? Hit us up at decoder@theverge.com. We really do read every email! *