Google Expands AI Video Tool to Cloud Customers
Veo is a new AI video generator invented by Google. You can create six-second 1080p video clips from text descriptions or images. Videos can be made with different styles and aspect ratios and they can be changed by masked editing and basic video physics simulation. Videos can contain cinematic effects, landscape shots, and time-lapse sequences. However, the tool still faces challenges, including inconsistencies in generated objects and issues with physics accuracy.
Veo competes with similar tools from companies like OpenAI, Adobe, and Meta, offering better performance and improved security. The model has been trained by including some YouTube content, in line with Google's data policies. Veo also includes features like content filters and watermarking to prevent illegal use, though some problems about deepfakes and copyright issues still remain unsolved.
Although it has a potential, Veo has limitations too. Tools like Veo could interrupt creative jobs. Google’s cautious approach to releasing Veo reflects its focus on improving the tool based on user feedback before making it widely available. It is not familiar when Veo will be fully available on Vertex AI or integrated with other Google services.