Google DeepMind Unveils Veo 2 AI Video Generator Surpassing OpenAI's Sora
Google DeepMind's Veo 2 AI video generator creates high-quality 4K videos up to two minutes long, outperforming OpenAI's Sora in resolution and duration.
Google DeepMind has unveiled Veo 2, a next-generation AI video generator that surpasses OpenAI's Sora in both resolution and video length capabilities. Veo 2 can create two-minute-plus clips in resolutions up to 4K (4096 x 2160 pixels), which is four times the resolution and over six times the duration of what Sora can achieve.
Key Features and Improvements
Veo 2 brings several advancements to AI video generation:
- Enhanced Realism: The model demonstrates an improved understanding of physics, camera controls, and produces clearer footage with sharper textures and images, especially in scenes with significant movement.
- Advanced Motion Capabilities: Veo 2 can more realistically model motion, fluid dynamics, and properties of light, including shadows and reflections.
- Cinematic Control: The AI understands cinematography concepts, allowing users to specify genres, lenses, and cinematic effects in their prompts.
- Reduced Hallucinations: Veo 2 produces fewer unwanted details or artifacts, resulting in more realistic outputs.
Availability and Integration
Currently, Veo 2 is exclusively available through Google's experimental video creation tool, VideoFX, where videos are capped at 720p and eight seconds in length. Google plans to expand access to more users and integrate Veo 2's capabilities into other products across the Google ecosystem.
Safety and Responsible Development
To mitigate the risk of deepfakes, Google is using its proprietary watermarking technology, SynthID, to embed invisible markers into frames generated by Veo 2. However, it's important to note that like all watermarking technologies, SynthID is not foolproof.
Comparison to Competitors
In head-to-head comparisons judged by human raters, Veo 2 has achieved state-of-the-art results against leading models.
Users who have gained access to Veo 2 report that its outputs are significantly more advanced than those of competitors, with one user describing the difference between Veo 2 and Sora as "comparing a bike to a starship".
Future Developments
Google plans to make Veo 2 available via its Vertex AI developer platform as the model becomes ready for use at scale. The company also intends to expand Veo 2 to YouTube Shorts and other products in the coming year. As AI video generation technology continues to advance, Veo 2 represents a significant leap forward in creating high-quality, realistic video content using artificial intelligence.

