On Tuesday, Snapchat introduced a range of new artificial intelligence (AI) tools during its 6th annual Snap Partner Summit. Among the highlights was the announcement of an AI video tool specifically designed for users with creator accounts. This innovative tool will enable users to produce videos using text and image prompts, with all AI-generated videos clearly watermarked by the company to distinguish them from standard content.
The announcement, outlined in a press release, emphasized the significance of the AI Video tool, officially named Snap AI Video. This feature will be exclusive to Creators on the platform. To qualify as a Creator, users must maintain a public profile, regularly post to their Stories and Spotlight, and successfully cultivate a substantial audience.
Functioning similarly to prevalent AI video generators, this feature allows for video creation from text prompts. Snapchat also plans to incorporate image prompt capabilities in the near future. Currently, the tool is available in a beta version on the web for a select group of creators.
A spokesperson for the company shared with TechCrunch that the AI video feature leverages Snap’s proprietary foundational video models. Once the feature is made widely accessible, Snapchat intends to implement icons and context cards that inform users when a Snap has been created using AI. The specific watermark will remain visible even if the content is downloaded or shared.
The spokesperson further mentioned that these video models have undergone thorough testing and safety assessments to ensure they do not generate harmful content.
In addition to the new video capabilities, Snapchat unveiled an AI Lens that allows users to visualize their older selves. Subscribers to Snapchat+ can now use Snapchat Memories with enhanced AI captions and Lenses. The platform’s native chatbot, My AI, is also receiving upgrades, enabling it to handle a wider array of tasks.
According to the company, users can now leverage My AI to tackle more sophisticated problems, such as interpreting parking signs, translating foreign menus, and identifying unique plants. Lastly, Snapchat is collaborating with OpenAI to provide developers access to multimodal large language models (LLMs), enabling them to create more Lenses that recognize objects and offer additional context.
