Rumor: OpenAI’s New ‘Audio-Based’ Device Could Arrive Very Soon

Screens had a good run. Now OpenAI is betting that the future of AI lives in your ears, not on your phone.

OpenAI is ramping up work on its audio AI as it prepares for an upcoming personal device that will rely primarily on voice, according to The Information. The device is expected to be “largely audio-based,” marking a shift away from screens as the main way users interact with AI.

Inside the company, researchers believe OpenAI’s current audio models are not yet on par with its text-based systems. While ChatGPT can already speak, the audio model behind that voice is separate and, according to current and former employees cited by The Information, it is slower and less accurate than the text version.

To close that gap, OpenAI has brought together teams from engineering, product, and research over the past two months to focus on improving audio AI.

A new audio-model architecture now in development reportedly delivers more natural and emotionally expressive speech, along with more accurate and detailed answers. The model is also expected to handle interruptions better and even speak at the same time as a user, something current systems struggle to do.

The new audio model is expected to be released in the first quarter of 2026, The Information reported.

A new generation of voice-first devices

OpenAI isn’t planning just one gadget, but a family of devices. The Information, citing multiple sources, says the discussed ideas include smart glasses and a screenless smart speaker.

Reports from multiple sources have also circulated claims that one form factor could be a pen-like device, integrating voice, handwriting, and ChatGPT into a single portable tool. While unconfirmed by OpenAI, it highlights the broad speculation around the company’s hardware ambitions.

This hardware vision is being shaped by former Apple design chief Jony Ive, whose firm, io, was acquired by OpenAI for nearly $6.5 billion earlier in 2025. Ive has signaled that reducing screen addiction is a personal priority.

The war on screens

OpenAI isn’t the only one trying to kill the screen. The tech industry is currently in a war on screens, with giants and startups alike betting that voice is the ultimate interface.

For instance, Meta recently upgraded its Ray-Ban smart glasses to act as a set of high-tech ears, helping users hear conversations clearly even in crowded, noisy rooms. That same hands-free logic is hitting the road, too, as Tesla begins tucking xAI’s “Grok” into its dashboards so drivers can adjust the climate or find a route just by talking.

Even the way we graze for information is changing, with Google testing audio overviews that skip the reading and turn traditional search results into easy-to-digest spoken summaries.

For OpenAI, the first of these devices is expected to land in about a year. The goal is to move away from apps and toward an ambient assistant that takes in the world through audio and video (when permitted) to proactively help users reach their goals.

For more on how OpenAI reached this moment, read our deep dive on OpenAI’s relentless rise and how the company reshaped the AI landscape at breakneck speed.
The post Rumor: OpenAI’s New ‘Audio-Based’ Device Could Arrive Very Soon appeared first on eWEEK.