Overview

• Launched on September 30, 2025, Sora 2 generates full videos from text by integrating image, motion, and sound with stronger physical realism and scene coherence.
• Access is limited to selected invitees in the United States and Canada, with broader availability planned gradually during 2026.
• Invitees currently use the service at no cost, and OpenAI has signaled future access tiers, including a possible Pro option to manage compute demands.
• New tools include cameos for adding a user’s voice or face, a community mode for remixing and sharing creations, and finer creative control over shot types and narrative pacing.
• Safety updates include strengthened moderation and provenance metadata designed to trace content origin and reduce misuse or misinformation.