OpenAI’s consumer hardware push begins in the living room. The company’s first device designed with Jony Ive is a mobile, screen-free smart speaker meant to sit in a home and behave like a humanlike companion, according to people familiar with the project who spoke to Bloomberg.
It surfaced the day before OpenAI’s actual first shipping hardware, a macro pad for Codex coders, reached customers.
The distinction matters for anyone counting firsts. Codex Micro is a developer accessory built with the boutique keyboard maker Work Louder, while the speaker is the first consumer product out of io, the startup OpenAI bought from Ive for $6.5bn last year.
Internally, OpenAI does not describe it as a speaker at all. It calls the product the first of its kind, a computer built for AI and aimed at making busy people more productive, one that controls smart-home appliances, plays media, answers questions, responds to messages, and reaches into the rest of ChatGPT.
The defining feature is meant to be personality. The device carries mechanical elements that can move on their own, engineered to create the sense that it is alive rather than an object waiting to be addressed.
It has a camera and other sensors to read its surroundings, and a rechargeable battery so it can be carried from the laundry room to the kitchen and later parked in a bedroom to play music. It will also draw on personal information such as emails to build a picture of whoever owns it.
The pitch, in other words, is proactivity. OpenAI wants the thing to anticipate needs and volunteer information rather than wait to be asked, growing more personalised as it learns its owner.
Voice is where the engineering shows. The speaker will run a more advanced version of GPT-Live, the voice mode OpenAI rolled out this month, which can listen and talk at the same time and adapt mid-conversation.
The stock market read it as a category threat rather than a curiosity. Sonos fell more than 10% in late trading before paring the losses, while Apple slipped less than 1% to a low of $313.52.
The io deal bought people as much as drawings. LoveFrom, Ive’s studio, is helping craft the lineup, Evans Hankey, once Apple’s head of industrial design, is leading development of the speaker, and OpenAI hired Vision Pro chief Paul Meade from Apple last month.
Apple has noticed. It sued OpenAI on 10 July for trade-secret theft, accusing chief hardware officer Tang Tan, a former head of iPhone product design and an io co-founder, of leading a campaign to obtain confidential information about Apple’s future products. The complaint counts more than 400 Apple staff who have since moved to OpenAI.
OpenAI’s position is that the speaker veers far enough from anything Apple currently sells that infringement is unlikely, and that it is not aware of any evidence the complaint has merit. Apple conceded in its filing that discovery would be needed to establish whether its technology is in fact being used.
The exposure is not abstract. Apple is seeking an injunction on OpenAI’s hardware, which could stall the company’s ability to sell anything at all.
Roughly five products are in development, with the speaker first. OpenAI is aiming to unveil it this year and release it in 2027, a schedule that now depends partly on a courtroom.
Further out, the company wants a mobile device capable of replacing the smartphone. It has explored wearables including a pendant, and has shown interest in home robotics.
Apple is building toward the same room. Its first AI home product, code-named J490, is a long-delayed command centre with a square seven-inch display, videoconferencing, facial recognition, and the rebuilt Siri arriving in iOS 27, and a version mounted on a robotic arm that repositions itself as you talk to it is also in the works.
An OpenAI spokesperson declined to comment on the speaker.
