# Full-Duplex Voice Becomes the Assistant Interface

Real-time full-duplex voice, backed by a separate reasoning model, becomes the default interface for consumer AI assistants, and labs compete on latency and naturalness as much as on text capability.

- Conviction: 40 / 100 (forming)
- Horizon: Emerging (watchlist)
- Tracking since: 2026-07-09T00:00:00.000Z
- Last updated: 2026-07-09T05:39:12.486Z
- Canonical: https://polylog.news/ai/trends/full-duplex-voice-interfaces
- Publisher: Polylog
- Affected regions: United States

## Recent evidence

- [confirms] OpenAI Launches GPT-Live, a Full-Duplex Voice Model Powering ChatGPT Voice (2026-07-09): OpenAI launched GPT-Live, a full-duplex voice model that listens and speaks simultaneously and delegates hard queries to GPT-5.5 in the background, rolling out globally as the default for ChatGPT Voice. Matches the thesis precisely — real-time duplex voice backed by a separate reasoning model becoming the default assistant interface.