Google Announced Its Most Powerful AI Image Generator Imagen 3
Google announced Imagen 3, a generative AI model for creating images that better understands text queries, makes fewer errors, and produces clearer text in images.
The head of Google’s Deep Mind AI research unit, Demis Hassabis, said during the presentation that Imagen 3 more accurately understands the test queries on which it creates images than the Imagen 2 model. He noted that the algorithm works more “creatively and detailed”, and also makes fewer mistakes and creates fewer “distracting artifacts”.
At the same time, Google tried to allay concerns that Imagen 3 will be used to create deepfakes. The image generation process will use SynthID technology, through which invisible cryptographic watermarks are applied to media files. This approach is expected to make attempts to use Google’s AI generator to create fake content useless.
Individual users can subscribe to Imagen 3 through Google’s ImageFX service. Developers and enterprise customers will have access to the AI generator through the Vertex AI machine learning platform.
