Unexpected Results: YouTube Shorts and Google Photos Use AI to Turn Photos into Videos
New AI Features in YouTube Shorts and Google Photos
YouTube Shorts and Google Photos have introduced a new feature that allows users to convert still images into video clips using generative AI. This update, part of several key enhancements released this week, brings the ability to create short video clips from photos on both platforms. Alongside this, there are additional AI-powered tools designed to enhance user experience.
Google Gemini already had the capability to generate videos from photos, which was rolled out earlier this month. Now, this functionality is being extended to YouTube Shorts and Google Photos.
Photo to Video Capability in Google Photos and YouTube Shorts
Both YouTube Shorts and Google Photos allow users to select a photograph and choose from pre-suggested prompt options. These prompts are not manually written by the user but are instead provided as suggestions.
In Google Photos, the feature starts with two options: “Subtle movements” or “I’m feeling lucky.” The company notes that photos featuring people tend to work best with this tool.
For YouTube Shorts, the feature offers the ability to add movement to landscape photos, animate everyday scenes, or transform group photos into videos.
Examples shared by Google highlight some common AI-generated effects. For instance, text on a flyer in a still photo may appear as gibberish in the video, and the image might flicker slightly. Additionally, elements like poles may not be perfectly aligned.
In Google Photos, the examples show animated scenes where people seem to be celebrating, such as throwing confetti or releasing balloons. A subtle warning appears: “Experimental GenAI: results may be unexpected.”
Josh Sassoon, Director of UX for Google Photos and Google One, mentioned in a blog post that these features are experimental and may not always produce the desired results. He encouraged users to provide feedback through thumbs up or down buttons, which will help improve the safety measures and overall experience.
AI Capabilities and Upcoming Updates
While Gemini uses Veo 3 to power its still-to-video tools, both Google Photos and YouTube Shorts currently use the older Veo 2. Google plans to introduce Veo 3 capabilities to YouTube Shorts later this summer.
In the US, the photo-to-video tool has already started rolling out on Android and iOS devices for Google Photos. For YouTube Shorts, the feature is set to roll out over the next week in the US, Canada, Australia, and New Zealand, with more regions expected to follow later this year.
Additional AI Features for YouTube Shorts
YouTube Shorts is also receiving an AI Playground, which includes new generative effects. These effects allow users to experiment with preset AI tools, such as transforming a walk down the street into a dive into a pool or turning drawings into images. Google states that these effects will be regularly updated.
The AI Playground is accessible via the Create section, where users can tap the sparkle icon. This new space will house AI creation tools, inspiration, and pre-written prompts, enabling users to generate videos, images, and music.
Google Photos Gets AI Remix and Create Tab
Google Photos will soon include an AI Remix tool, which generates versions of photos based on pre-selected styles, such as comics or sketches. This feature is scheduled to roll out for US users in the coming weeks.
Starting in August, Google Photos will also introduce a dedicated Create tab. This tab will house new generative AI tools alongside options for creating collages and highlight videos.
Google emphasizes that both Google Photos and YouTube Shorts will include a SynthID watermark to identify AI-generated content.
Conclusion
These updates reflect a growing trend of integrating AI into everyday applications, enhancing user creativity and engagement. As these features continue to evolve, they offer new possibilities for how users interact with their media.