February 9 news.OpenAI FIRST PARAGRAPH FOR CONSUMER AIHardwareIt's accelerating, but the first version of the moonlight this September will be a limited "simplified" version。
The reason for this is that HBM supply is strained to raise the cost of 2nm chips, forcing OpenAI to postpone the "abnormal form" of the calculation unit in the original plan by launching a version that only supports audio functions。
The blogger's latest article, "Intelligence Picachu", states that the device is namedDimeIt's small in size。
Its patent was published yesterday at the National Intellectual Property Administration of the United States, with metal-based materials and pebble-like objects, with two desirable capsule headsets inside, wearing them behind their ears。
Supply chain information indicates that the equipment is more closely related to the cell phone level, that the primary processor is targeted to a 2nm smart phone chip and that customized chips are being developed to achieve direct implementation of the iPhone Siri command through voice。
Inside OpenAI, the equipment, code-named Sweetpea, was given the highest priority by the Johnny Ive team, with a target of between 40 and 50 million for the first year. Fuscon has also been informed of the need to prepare the OpenAI 5 equipment for production by 2028。
OpenAI CEO Sam Ottman has publicly stated that:The real competition is not Google, but Apple。
In his view, the main battlefield of the future AI is at the terminal, not at the cloud end; smartphone screens and interactive means limit the potential of the AI partner, so OpenAI must build "AI Primary Devices"。
Ottman compared his vision to the "Hav by Lake" — providing a focus for users outside the Times Square of information bombing。
Besides headphones, a mysterious AI pen is also being developed. In combination with the leads mentioned repeatedly by Altman and Johnny Ive, it is assumed that the equipment is small, environmentally sensitive and may use high-quality sensory materials such as ceramics, with very simple interactions at its core。
At the technical level, OpenAI is accelerating the iterative audio model, laying the foundation for hardware. Informed sources revealed that the new generation of models, which not only make the voice more natural, but also support synchronized dialogue and interruption processing, is expected to be released in the first quarter of this year。
OpenAI has formed teams across the supply chain, industrial design and model development, with the goal of creating "smart partners" that can work with initiative rather than simple voice interfaces。
IT IS ALSO ASSUMED THAT THE AI PEN MAY BE INTEGRATED INTO A MICRO PROJECTOR, WHICH PROJECTES THE IMAGE TO THE DESKTOP TO SOLVE THE PROBLEM OF NON-SCREEN INTERACTION, AND THAT THE FOLDER MAY BE INTEGRATED INTO A MICROPHONE OR CAMERA TO ACHIEVE TEXT RESOLUTION AND ENVIRONMENTAL AWARENESS。
WHEN USERS WRITE ON PAPER, AI CAN INTERPRET CONTENT IN REAL TIME, GENERATE TO-DOS, OR EVEN CONTROL THE PERIMETER AS A SMART HUB。