Generative AI moves from chat tools to video, devices and governance

Generative AI is moving beyond automated text into video, audio, cloud infrastructure and legal governance. OpenAI’s Sora app service ended on April 26, 2026, with API service ending on September 24 of the same year, while newer tools such as Google’s Gemini Omni, MiniMax’s H3, Kling and Runway are reshaping video production with multimodal capabilities.

The latest models are positioned as practical tools for business users, not only engineers or video specialists. Improved handling of images, audio and text could make short product explainers, internal training clips, marketing tests and presentation videos faster and cheaper to create from prompts.

Canon launched a cloud-based functional expansion platform on October 1, 2026, combining its imaging and optical technologies with generative AI. The platform points to a model in which devices such as cameras, scanners and office equipment gain new cloud-delivered features for image analysis, automation and workplace DX after purchase.

Legal and governance questions are becoming more urgent as AI voice tools spread in training, multilingual support and customer service. A lawsuit over TikTok videos that allegedly imitated real voice actors’ narration ended with the voice actors’ claims dismissed, underscoring the need for companies to confirm commercial rights, avoid unauthorized imitation and disclose AI-generated audio or video use.