OpenAI plans to announce a new audio language model in the first quarter of 2026, according to a report in The Information. The model is intended as a deliberate step toward an audio-based physical device. The company has combined teams across engineering, product, and research into a single initiative focused