While the AI video world awaits OpenAI's Sora with bated breath, Google is quietly executing a masterstroke in accessibility and control, bringing a powerful new feature to its Veo model that could change how we create video content.
The generative AI video landscape is heating up, but it's a fragmented battlefield. For creators and businesses in Europe, the highly-anticipated OpenAI Sora 2 remains conspicuously out of reach, leaving a significant gap in the market. Seizing this opportunity, Google is aggressively refining its competitor, Veo, and with its latest move, it's addressing one of the most significant pain points in AI video generation: precise creative control.
Following the launch of Veo 3.1 in mid-October, Google has now integrated its flagship video model more deeply into the Gemini app for both mobile and desktop users. But the real news isn't just the integration; it's a powerful new capability that allows users to upload multiple reference images alongside their text prompts, finally giving creators the tools to steer the AI's vision with unprecedented accuracy.
From Vague Prompts to Precise Direction: How Multi-Image Input Works
Until now, generating an AI video has often been a game of chance. A text prompt like "a knight walking through a enchanted forest at sunset" could yield a thousand different interpretations, rarely matching the specific picture in the user's mind. Google's new feature for Veo shatters that limitation.
As announced on the official X account for the Gemini app, users can now upload several images to guide different aspects of their video.
Here’s a practical example of how this transforms the creative process:
- Image 1 (The Character): You can upload a detailed drawing of your original character—a specific knight with unique armor and a crest.
- Image 2 (The Background): You can then upload a photograph of a misty, ancient redwood forest to define the setting.
- Image 3 (The Style): Finally, you could add a still from an animated film like Princess Mononoke to dictate the overall visual aesthetic and lighting.
Your text prompt then becomes the director, instructing the AI on how to combine these elements: "Animated in the style of reference image 3, show the knight from reference image 1 cautiously walking through the forest from reference image 2 as the sun sets."
This multi-layered approach moves AI video from a novelty to a genuine tool for storytellers, designers, and marketers who need consistency and brand alignment.
The official announcement was made directly by the Gemini team, showcasing the feature's potential.
Democratizing AI Filmmaking: From Niche Tools to Mainstream Apps
It's worth noting that this "Ingredients to Video" feature isn't an entirely new concept from Google. It was first introduced in October within Google's specialized AI filmmaking tool, Flow, and is also available on their developer-centric Vertex AI platform.
However, its integration into the main Gemini app is the real game-changer. This strategic move demystifies and democratizes a powerful filmmaking technique, taking it out of the realm of specialized developer tools and placing it directly into the hands of everyday users. You no longer need to be an AI engineer or a professional filmmaker to leverage this level of control; you just need the Gemini app on your phone or computer.
This focus on accessibility, especially while its main competitor is not yet available in a major market like Europe, gives Google a significant strategic advantage. It's not just about having a powerful model; it's about building a user base and fostering creator loyalty through superior and more intuitive tools.
The Road Ahead for AI Video
According to Google, the rollout of this new multi-image feature within the Gemini app has already begun. While the AI video generation space is still in its relative infancy, features like this signal a clear direction for the future: more control, more consistency, and a tighter feedback loop between human intention and machine output.
For European creators feeling left behind in the Sora waitlist, Google Veo's latest evolution offers a compelling and increasingly sophisticated alternative. The race for AI video supremacy is far from over, but by focusing on user-centric features and broad accessibility, Google is proving it's a serious contender ready to win over the hearts and minds of creators worldwide.
What do you think? Does multi-image control address your biggest frustration with AI video? Let us know in the comments below.
We’re back with another update to Veo 3.1:
— Google Gemini App (@GeminiApp) November 14, 2025
Rolling out now on mobile and desktop, you can upload multiple reference images alongside your video prompts, to create entirely new worlds and more nuanced videos that are true to your vision.