OpenAI plans its first hardware product as a donut-shaped speaker without a screen, as a report from the news agency Bloomberg from August 6 shows. The device is expected to cost over 300 dollars and, according to the schedule, will not be available until 2027. At the same time, OpenAI is seeking to dismiss Apple’s lawsuit over allegedly stolen trade secrets.
Speaker gets metal body and light ring for listening mode
The device is said to be about the size of a hockey puck, made of metal, and features a donut-shaped indentation in the middle to facilitate one-handed carrying. Moving parts are intended to make the speaker feel more lively than traditional smart speakers – a feature that was already known from a previous report on the project. New additions include a built-in camera, multiple microphones, environmental sensors, and a lighting system that signals when the device is listening.
OpenAI positions the product internally as a kind of AI computer that is operated solely by voice and does not require a screen. The built-in models go beyond the capabilities of the existing ChatGPT voice mode and adapt over time to the habits of users. Information regarding price, shape, and functionality comes exclusively from Bloomberg’s report, which cites unnamed sources familiar with the matter and has not been independently verified.
Launch date slips towards 2027, price remains unclear
Originally, OpenAI had aimed to unveil the device in 2026, with sales starting in 2027. Meanwhile, the unveiling itself is also moving closer to 2027, without OpenAI confirming a specific date. Various reports suggest a price range between 300 and 400 dollars; there is currently no conversion to euros or information regarding availability in Germany and the EU. The distribution method – whether through a dedicated OpenAI store or established electronics retailers – is also open.
According to the company, the speaker is just one of about five tested hardware concepts, which also includes smart glasses. OpenAI consciously chose the speaker as its first product because it raises fewer privacy concerns compared to body-worn cameras. With the Codex Micro introduced in July, the company is already selling its first physical accessory product for developers – the speaker would be the first standalone consumer device.
OpenAI seeks dismissal of Apple lawsuit
The product plans coincide with an ongoing legal dispute with Apple, which filed a lawsuit against OpenAI in July over allegedly stolen trade secrets. On August 6, OpenAI submitted a 31-page motion to dismiss, according to TechCrunch. Apple has neither adequately described which information is to be considered a trade secret nor plausibly articulated a legal claim to it, OpenAI’s lawyers argue. The motion states that OpenAI is building something completely new and different from anything at Apple.
OpenAI also accuses Apple of overlooking its own security failures: the company allegedly allowed former employees to use private iCloud accounts for work purposes and did not consistently revoke access after their departure. A federal judge in Northern California is set to hear the motion on October 1. Until a decision is made, it remains unclear whether a temporary injunction could further delay the market launch.
It will be crucial whether the motion to dismiss is successful before OpenAI enters the critical phase of actual product manufacturing. If the motion fails, a discovery process with access to internal hardware documents could further strain the already tight schedule leading up to 2027.
