If you have an Amazon Echo on your counter or a Google Nest Hub in the kitchen, you have already lived through the first generation of home AI speakers. Alexa can set a timer, play a song, report the weather. Ask it a follow-up question and it forgets the conversation happened. Say something while it's mid-response and it doesn't register you at all. The voice feels robotic. The interaction feels like navigating a phone tree. That's the gap OpenAI is now going after.

The Information reported this week that OpenAI's first consumer hardware device is priced at over $300, with Bloomberg's sources pointing to a $300–$400 range for a 2027 release. The design is from Jony Ive's firm LoveFrom — that's the designer behind the original iPhone, the iPod, the MacBook Air. The device is reportedly ring-shaped, roughly the size of a hockey puck, and battery-powered. And it has moving parts.

That last detail I'll come back to.

What's actually different about this

Most home AI speakers work in what's called turn-taking voice. The device waits for a wake word. You say your command. It processes. It responds. Then it goes back to waiting. Each exchange is one complete turn — like a walkie-talkie where you have to finish speaking before the other person can respond.

OpenAI's device would be built around a different kind of voice conversation entirely. Full-duplex voice means the AI is listening and speaking at the same time, the same way a phone call works between two people. If you interrupt it mid-sentence, it registers that immediately. If you say "wait, actually I meant—" it hears you in the moment and adjusts. That's not how Siri or Alexa work. They need to finish before they can hear you again.

This matters more than it sounds. The awkward, stilted quality of conversations with current home speakers isn't just that they give bad answers. It's that the rhythm of the conversation is wrong. You feel like you're issuing commands, not talking. Full-duplex voice changes the conversational feel at the infrastructure level — not just with smarter answers but with a more natural back-and-forth.

Object lesson: remember when Siri launched in 2011, and everyone's first impression was "impressive demo, frustrating in practice"? The frustration came mostly from the turn-taking constraint — you said a thing, Siri thought, Siri answered, and that was it. OpenAI has been working on closing that gap since it launched GPT-Live in July. This device would be the first physical home object built around that technology.

The Jony Ive factor

In case the name isn't familiar: Jony Ive spent 27 years at Apple and is largely responsible for what iPhones, iPods, and MacBooks look and feel like. His design firm LoveFrom launched after he left Apple. He and OpenAI CEO Sam Altman announced a collaboration on a family of AI hardware products — this speaker appears to be the first of them.

Why does this matter for a speaker? Because Ive's design philosophy isn't "make it look cool." It's "make it feel like it belongs in your life." The original iPod didn't win because it was technically superior to other MP3 players. It won because the physical experience — the scroll wheel, the weight, the way it fit in your palm — made the technology feel natural rather than complicated. If anyone can make an AI speaker that feels like it belongs on your counter rather than sitting there as an obvious piece of technology, Ive's track record suggests he might be the one.

About those moving parts

TechCrunch reported the device will have components that physically move, described as intended to make it more lifelike and engaging.

Two plausible interpretations. The optimistic one: the device turns toward you when you speak, the way a person does when they're listening — a physical signal that it's paying attention. That's not gimmick. That's actually good design. Some high-end video conferencing equipment already does directional tracking. A speaker that physically orients toward your voice is a more intuitive signal of "I'm listening" than a glowing ring.

The more skeptical interpretation: it's an attempt to give the AI physical expressiveness — body language for a machine — and that kind of design can land somewhere uncomfortable. There's a term in robotics called the "uncanny valley," which describes the strange feeling you get when something is almost-human but slightly wrong. A speaker with moving parts that tries to emulate animal or human expressiveness might land there.

I can't tell which from the description alone. That's the part you have to see to judge.

Source spread

• The Information — ChatGPT speaker will cost over $300 [hype] — first to publish price; framed as confirmed by sources close to the product
• Engadget — Ring-shaped smart speaker reportedly $300–$400 [builder] — added the ring/hockey-puck form factor and the Jony Ive LoveFrom sourcing
• TechCrunch — OpenAI's new AI smart speaker [builder] — moving parts detail; battery-powered and portable framing; noted 2027 release
• TechTimes — ChatGPT Speaker With Moving Parts: What GPT-Live Architecture Makes Possible [hype] — editorial analysis connecting the device's design to GPT-Live voice capabilities; framing is enthusiastic

What's real / What deserves a side-eye

What's real:

• Full-duplex voice is a genuine improvement over first-gen home speakers. The conversational rhythm difference is real — not just a benchmark gap but a felt difference in daily use.
• Jony Ive's involvement is meaningful. They went to the most prominent industrial designer alive. That signals product seriousness, not an internal team doing a quick hardware spin-out.
• Battery-powered and portable is a genuine differentiator. Every Amazon Echo and Google Home needs a power cable, which means it stays in one place. A device you can carry between rooms or take outside is a different use case entirely.

What deserves a side-eye:

• This is all from sources, not an official announcement. OpenAI has not announced this product. Price, specs, and timeline could change between now and 2027.
• The $400 price is a difficult sell without proof. Amazon and Google both tried to sell premium home audio (Echo Studio at $200, Nest Hub Max at $230) and hit resistance. $400 for a speaker in a category where $50 incumbents exist requires dramatically better experience to justify.
• "More lifelike and engaging" can mean many different things, and without seeing the actual device it's impossible to know if the moving parts land as Ive intends or as their own category of weird.

What to do about it

• Don't act on this yet. There is no official product announcement. No product page. No preorder. Everything here is from sources. Wait for OpenAI to actually announce the device before making any decisions.
• If you're shopping for a home speaker now, the existing options are fine. Amazon Echo and Google Nest are mature products that do what they do reliably. There's no reason to wait for a 2027 device if you need something today.
• When reviews come out in 2027, the thing to test specifically is the conversation rhythm. Not "did it give a correct answer?" but "did it feel like talking to something, rather than issuing commands?" That's the differentiator this device is betting on. If reviewers say the full-duplex voice feels natural in a home environment, that's the signal.
• Watch the moving parts closely. That detail will either be praised as inspired design or quietly removed before launch. Either outcome is informative about what OpenAI is actually trying to build.

Further reading

• The Information — ChatGPT speaker will cost over $300 — primary source on price
• Engadget — ring-shaped smart speaker reportedly $300–$400 — form factor and LoveFrom details
• TechCrunch — OpenAI's new AI smart speaker — moving parts and portability framing
• OpenAI — Introducing GPT-Live — the full-duplex voice system that would power this device