Something that moves feels like a companion, not a tool.
In a moment that marks the convergence of artificial intelligence and domestic space, OpenAI is preparing to cross the threshold from software into the physical world — building a screen-free smart speaker designed not merely to answer, but to inhabit. The move arrives shadowed by a lawsuit from Apple, which alleges that OpenAI's aggressive recruitment of former employees was less about talent and more about secrets. What unfolds now is not simply a product launch, but a contest over who will shape the intimate architecture of the AI-first home.
- OpenAI's first consumer hardware device — a mobile, screen-free smart speaker with mechanical movement, a camera, and the ability to learn its owner through email access — signals an ambition to make AI feel genuinely present in the home.
- Apple has filed suit against OpenAI, alleging that the systematic hiring of over 400 former Apple employees, including key design leaders, was a deliberate strategy to extract confidential product knowledge and accelerate hardware development.
- OpenAI's chief hardware officer Tang Tan, former iPhone design lead, sits at the center of the legal complaint, alongside Evans Hankey and Paul Meade — a concentration of Apple's most senior design talent now building for a direct competitor.
- OpenAI insists its device is meaningfully distinct from Apple's HomePod line and that it has no interest in trade secrets, but an injunction could still derail a planned unveil this year ahead of a 2027 release.
- The broader hardware race is accelerating rapidly — Apple is developing its own AI home hub with a robotic arm, AI startup Hark just raised $700 million, and Sonos shares fell over ten percent on news of OpenAI's entry into the market.
OpenAI is preparing to move from software into the living room. Its first consumer hardware device is a portable, screen-free smart speaker fitted with mechanical parts that shift on their own, a camera, sensors, and GPT-Live — the company's upgraded voice model capable of simultaneous listening and speaking. The device is designed to feel present rather than merely responsive, learning its owner over time, moving between rooms on battery power, and managing smart home functions without ever displaying a screen.
The announcement arrives under legal cloud. Apple has filed suit against OpenAI and two former employees, alleging that the company's hiring practices amounted to a systematic effort to extract confidential product information. At the center of the complaint is Tang Tan, OpenAI's chief hardware officer and former leader of iPhone product design, accused of orchestrating efforts to obtain sensitive Apple engineering details. Also named are Evans Hankey, now overseeing the speaker's development, and Paul Meade, a veteran of the Vision Pro headset recruited just last month. OpenAI has hired more than four hundred former Apple staff in total, according to the lawsuit.
OpenAI has responded with measured confidence, arguing its device differs substantially from anything Apple sells and that it has no interest in competitors' trade secrets. Still, the litigation carries real risk — an injunction could delay the speaker's planned unveil this year ahead of a 2027 release.
The speaker is one piece of a far larger bet. OpenAI is developing roughly five devices in total, backed by Jony Ive's design studio LoveFrom following a six-and-a-half-billion-dollar acquisition of io Products. A smartphone, wearables, and home robotics are also in development. Apple, meanwhile, is building its own AI home hub — a seven-inch display device with facial recognition and a robotic arm, set to run the new Siri arriving with iOS 27. What was once Amazon and Google's domain is becoming a crowded and consequential contest, with the definition of the AI-first home very much still up for grabs.
OpenAI is about to walk into the living room. The company, known for building software that talks, is now building something that moves—a portable smart speaker without a screen, fitted with mechanical parts that shift on their own, a camera, sensors, and the ability to learn who you are by reading your emails. It's being developed as what the company calls a humanlike AI companion for the home, a device that feels present rather than merely responsive. The speaker will run on GPT-Live, OpenAI's upgraded voice model that listens and speaks at the same time, adapting fluidly as conversation unfolds. It can move between rooms on a rechargeable battery, handle smart home controls, and tap into the full range of ChatGPT's capabilities—all without a screen to look at.
The timing, however, is fraught. Days before news of the device leaked, Apple filed a lawsuit against OpenAI and two former Apple employees, alleging that the company has systematically acquired confidential Apple information through hiring practices and supplier relationships to accelerate its hardware ambitions. Apple described the allegations as merely the beginning, suggesting more could surface during discovery. At the center of the complaint is Tang Tan, OpenAI's chief hardware officer and former leader of iPhone product design at Apple, who is accused of driving efforts to obtain sensitive details about Apple's future products and engineering methods. Alongside him are Evans Hankey, Apple's former head of industrial design now overseeing the speaker's development, and Paul Meade, a longtime Apple executive who previously led development of the Vision Pro headset and was recruited last month. In total, OpenAI has hired more than four hundred former Apple staff, according to the lawsuit.
OpenAI's response has been measured. The company maintains that its forthcoming device differs substantially from anything Apple currently sells, including the HomePod and HomePod mini, and says it is unlikely to breach any trade secrets. "While we take these allegations seriously, we're not aware of any evidence that this complaint has merit," a spokesperson said, adding that the company has "no interest in other companies' trade secrets" and believes in fair competition and the freedom for people to work wherever they choose. The legal battle could still disrupt the timeline: OpenAI is aiming to unveil the speaker this year ahead of a planned 2027 release, but an injunction from Apple could delay that.
The speaker is just one piece of a much larger hardware ambition. OpenAI is developing roughly five devices in total, built with support from Jony Ive's design studio, LoveFrom, following OpenAI's six-and-a-half-billion-dollar acquisition of io Products. Beyond the speaker, the company is exploring a smartphone developed with component makers including MediaTek, Qualcomm, and Luxshare, along with wearables and home robotics. The scale of the bet is significant, and the market is watching closely. Shares in Sonos fell more than ten percent in late trading before recovering some ground. Apple's stock dipped slightly to around $313.52.
Apple, for its part, is not sitting idle. The company is developing its own smart home command centre, codenamed J490, featuring a square seven-inch display, video calling, facial recognition, and a version mounted on a robotic arm that can reposition itself. It will run around the new Siri assistant arriving with iOS 27. Elsewhere in the market, investor appetite for AI hardware is already evident. Hark, an AI hardware startup founded by Brett Adcock, recently closed an oversubscribed $700 million funding round at a $6 billion valuation. What was once the domain of Amazon and Google is now becoming a crowded field, with OpenAI and Apple both racing to define what an AI-first home device should be—and who gets to build it.
Citas Notables
While we take these allegations seriously, we're not aware of any evidence that this complaint has merit. We have no interest in other companies' trade secrets and believe in fair competition and allowing people the freedom to work wherever they choose.— OpenAI spokesperson
Apple described the allegations as merely the tip of the iceberg, suggesting further findings could emerge during discovery.— Apple's lawsuit filing
La Conversación del Hearth Otra perspectiva de la historia
Why does OpenAI need to make a physical object at all? Isn't the software enough?
Because software lives in someone else's device. A phone, a speaker, a computer—those are all controlled by other companies. If OpenAI wants to own the relationship with the user, it needs to own the hardware too. The speaker is the anchor.
And the moving parts—the mechanical elements—what's that about?
Presence. A device that just sits there and talks feels like a tool. Something that moves, that shifts its orientation toward you, that has a body—that feels like a companion. It's the difference between talking to a box and talking to something that's alive.
Apple's lawsuit seems serious. Do you think OpenAI actually stole trade secrets?
The lawsuit alleges systematic hiring of Apple's best people in key roles. Whether that's theft or just smart recruiting depends on what those people actually did with what they knew. The legal system will sort it out. But the optics are real—Apple is saying OpenAI used insider knowledge to move faster.
What happens if Apple wins an injunction?
The speaker doesn't ship this year. OpenAI's entire hardware timeline gets disrupted. And it sends a signal to other companies that hiring talent from competitors comes with legal risk.
Why is Sonos stock falling?
Because if OpenAI's speaker is as good as it sounds, and it's integrated with ChatGPT, it's a threat to every standalone smart speaker company. Sonos makes speakers. OpenAI is about to make a speaker that's also an AI. That's a different category entirely.
What does Apple's J490 tell us?
That Apple sees this coming and is building its answer. A screen, a robot arm, Siri—it's Apple's way of saying we're not ceding the home to OpenAI. The real competition isn't about speakers. It's about who owns the interface between you and AI.