Google Veo and Imagen 3 aim at OpenAI’s Sora, DALL-E

Google’s AI adoption coincides with OpenAI’s rapid global popularity. Moreover, Google’s new AI tools seem like counterparts to OpenAI’s originals.

OpenAI showed the world that AI is more than text with its AI image creator, DALL-E. Recently, it teased a text-to-video tool named Sora.

READ: OpenAI shows content made with Sora

In response, the search engine company unveiled its answer to these tools: Google Veo and Imagen 3.

How do Google Veo and Imagen 3 work?

Google Veo is an AI tool that generates videos from text descriptions. The search engine firm says Veo is its “most capable video generation model to date.” 

“It generates high-quality, 1080p resolution videos that can go beyond a minute, in a wide range of cinematic and visual styles,” the official Google DeepMind page adds.

Previously, AI videos were a laughing stock due to a famous meme featuring Will Smith. It shows the Hollywood star in a deformed state, eating pasta with his hands. 

However, Google DeepMind’s examples are stunning and have none of the telltale signs of AI-generated clips. 

The examples include “A lone cowboy rides his horse across an open plain at a beautiful sunset, soft light, warm colors.” 

Google says it will bring some of Veo’s features to YouTube shorts and other products. On the other hand, Imagen 3 is a text-to-image model.

Write a description of your desired picture, and Imagen will create it. Google says the AI model “better understands natural language, the intent behind your prompt and incorporates small details from longer prompts.”

You may try these Google Veo and Imagen 3 early by joining Google’s waitlist. However, it will only choose select online creators to test these AI programs.