Google has launched Veo, its latest generative AI video model, available for businesses to integrate into content creation workflows.
First revealed in May, Veo has surpassed OpenAI’s Sora, which was introduced three months later, by hitting the market through a private preview on Google’s Vertex AI platform.
Veo can generate high-quality 1080p videos in various cinematic and visual styles from text or image prompts.
While initial demos indicated clips could last “beyond a minute,” Google has not set a specific length limit for this preview release.
New sample videos released by the company are nearly indistinguishable from real footage, showcasing the model’s impressive capabilities.
Google’s Imagen 3 text-to-image generator is expanding to all Google Cloud customers via Vertex starting next week. Initially available through Google’s AI Test Kitchen in the US in August, the latest version now offers additional features.
Users on the allow list can access prompt-based photo editing and customize generated images by incorporating their brand, style, logo, subject, or product features.
Starting next week, Google’s Imagen 3 text-to-image generator will be accessible to all Google Cloud customers via Vertex, following its initial launch in the US through Google’s AI Test Kitchen in August.
Users on Google’s allow list will also gain access to new features, including prompt-based photo editing and the ability to personalize generated images by incorporating their own brand, style, logo, subject, or product features.
Google’s Veo and Imagen 3 come with built-in safeguards to prevent harmful content generation and protect against copyright violations, though the latter has been relatively easy to bypass.
Both models incorporate DeepMind’s SynthID technology, an invisible digital watermark that Google claims helps reduce misinformation and misattribution issues.
This concept is similar to Adobe’s Content Credentials, which embeds similar protections in content created using Adobe’s generative AI tools.
With Google’s video model now available, OpenAI is falling behind, and its Sora release, promised by the end of 2024, is looking less certain.
AI-generated content is already being used in campaigns like Coca-Cola’s recent holiday ads, and businesses have strong incentives to move forward without waiting for Sora, especially considering that 86 percent of companies using generative AI are seeing revenue growth, according to Google.
Stories You May Like