Google has started implementation of private access to its Veo and Imagen 3 generative AI models. Starting today, customers of the company’s Vertex AI Google Cloud package can start using it I see to generate videos from text instructions and images. So, starting next week, Google will make Imagen 3, its latest text-to-image framework, available to the same users.
With the launch of Veo, Google claims to be the first hyperscale cloud provider to offer an image-to-video model. At that point, OpenAI Sora The template is still only available to select artists, academics, and researchers, though that could change quickly with the company anticipating 12 days of product demos starting December 5th
Of Veo, Google says the model creates “consistent and coherent” 1080p footage and that it can run “longer than a minute.” The tool is also capable of working with both text and image instructions. In the latter case, you can use AI-generated or human-created images as the starting point for a video.
Looking at the example footage shared by Google, it’s clear that Veo, like all AI models, can struggle with cause and effect. For example, in the roasted marshmallow clip, the treats do not yellow or char when exposed to the heat of a campfire flame. Artifacts are also a problem, as is evident if you look closely at the hands in the concert footage.
As for Imagen 3, Google says the model generates “the highest-quality, most realistic images from simple text instructions, outperforming previous versions of Imagen in terms of detail, lighting, and artifact reduction.” Again, though, you don’t have to look too closely to see that Google still has a lot of work to do.
In the first example of a group of friends sitting on the trunk of a car, the original tip includes mention of “flash photography,” but the subjects are clearly backlit. You could argue that a flash was used to create intense backlighting, but if the idea behind the suggestion was to create something representative of 1960s flash photography, this image isn’t the one.
However, Google wants more enterprise customers to use generative AI. Citing your own researchsays the tech giant, among companies using generative AI in manufacturing, 86% report an increase in revenue. However, a recent Appen survey found that the return on investment from AI projects decreased by 4.6 percentage points from 2023 to 2024.
If you purchase something through a link in this article, we may earn a commission.
