Voice Input Beats Typing When Starting with AI

Ethan Mollick recommends beginning AI sessions in voice mode through the Gemini app or ChatGPT app and website. These two platforms currently deliver the strongest voice implementations, while Claude's version lags behind in quality and responsiveness. Users simply speak their prompts instead of typing, which lowers friction and speeds up iteration.

This technique shifts users from formal written prompts to natural conversation, changing how they experiment with AI. It reduces the intimidation of prompt engineering and encourages quick testing of ideas. The result is faster learning and less hesitation before each session.

Ethan Mollick demonstrates this workflow in his One Useful Thing newsletter, where he tests voice-first approaches across multiple models. His readers report quicker idea generation and fewer abandoned prompts when they switch to speaking rather than typing.