OpenAI is gearing up to launch a major overhaul of its audio artificial intelligence, introducing a powerful new model that can actually speak at the exact same time as a person. This isn’t just a minor tweak, it’s a huge leap toward natural conversation, and the company is aiming to release this model in the first quarter of 2026, as per The Information.
This massive effort is the essential groundwork for a whole new family of audio-first personal devices they plan to roll out to consumers. We’re talking about wearables and other gadgets designed specifically for AI interaction. While the first of these devices isn’t expected for about another year, the company is already unifying its engineering, product, and research teams to ensure the underlying audio models are ready for prime time.
Right now, if you talk to ChatGPT, the large language model running the audio response is separate from the model handling the text-based responses. Frankly, the audio side is lagging behind. According to employees, the current audio models aren’t as accurate or fast as the text answers, which is awful for trying to have a real-time conversation.
The new audio-model architecture is already paying off, though
The responses are reportedly more natural and sound emotive, plus they offer much more in-depth and accurate information. The best part, which I think is a top-tier feature, is that ability to speak simultaneously with a user and handle interruptions gracefully. That makes the interaction feel so much more human than current models allow.
OpenAI is joining other tech giants, including Google, Amazon, Meta Platforms, and Apple, in the race to develop next-generation AI devices. Many of these companies believe that today’s most popular gadgets, like the iPhone, aren’t optimized for the future of AI technology.
The design philosophy for these future gadgets is truly compelling: researchers want users to interact purely through speech, ditching the screen entirely. The idea is that speaking out loud is the most natural way to communicate, since that’s primarily how people interact with each other. By removing the display, they also hope to reduce the chances of device addiction, which has brought bans from regulatory bodies recently.
Even former Apple design chief Jony Ive, who is working with OpenAI on this hardware, sees the screen-less approach as a priority. Ive has been quite blunt about owning up to past tech consequences, stating, “Even if you’re innocent in your intention, I think if you’re involved in something that has poor consequences, you need to own it. That ownership, personally, has driven a lot of what I’ve been working on.”
These devices won’t just be conduits for apps, either. Researchers told staff that the device will act like a true companion, proactively offering suggestions to help users meet their goals. It’ll be able to process audio and video information about its surroundings, provided the user allows it.
OpenAI isn’t developing just one device, but a whole family of products they plan to release over time. Among the ideas they’ve discussed are AI-powered glasses and a smart speaker that doesn’t include a display. It’s an all-hands effort to make audio the next big thing in AI interaction, following its massive tie-up with Disney.
Published: Jan 5, 2026 02:00 pm