The update enables users to dictate intricate instructions, such as debugging code or managing pull requests, through a single vocal prompt. On macOS, the integration goes further by utilizing Appshots, which allows the model to interpret visual data from the user's screen. The system maintains responsiveness during these interactions, capable of handling interruptions and requesting clarification mid-process. While the mobile version of the feature previously focused on conversational fluidity, this desktop iteration prioritizes functional agency, allowing the software to navigate local applications and websites independently.
This release places OpenAI in direct competition with Anthropic, which recently upgraded its own voice mode for Claude. Anthropic’s updated interface now interacts with productivity platforms including Gmail, Slack, and Notion via its Opus, Sonnet, and Haiku models. OpenAI’s shift toward voice-driven computer use suggests a broader industry push to move AI assistants beyond simple text generation and into the role of active, desktop-integrated operators.
Comments (0)
No comments yet. Be the first!