Microsoft has begun rolling out Image Creator from Microsoft Bing in preview in select markets, and is preparing the AI art generator for a wider rollout to Microsoft Edge later this month. In a blog post and accompanying video, the company demonstrated how Image Creator will work and explained in more detail what restrictions it will place on user-generated challenges.
Microsoft said last week that it would bring AI art to both Bing and Edge, using the more advanced DALL-E 2 algorithm to create the art. It looks like Image Creator will be accessible from Bing.com and a related version will be available from Edge soon. The company showed off Image Creator working in the Edge sidebar, which can carve out a small vertical column to display search results and other information, as well as some useful tools. Here you will have access to the new image creator.
In the video, Microsoft showed how users can generate a prompt using conventional expressions such as art styles. In the video below, you can see that Image Creator returns four small results in a few seconds. It is unclear whether this will be representative of general performance. It’s also unclear if there will be some kind of credit system or other meter to limit prompt generation.
Likewise, Microsoft showed Image Creator running in Edge as well.
Here, Microsoft’s approach is more social: the example shown is a user who uses Image Creator’s content creation tools to create a “dream house” concept and then share it on social networks. Again the image appeared within seconds and four images were generated.
Microsoft’s blog post indicated that the AI art generation tools will work similarly to other services such as Midjourney or DreamStudio running on the Azure cloud. “We’ve found that, in general, Image Creator works best when you enter a description of something with additional context, such as the location or art style you want to emulate, as opposed to a more limited description,” Microsoft said.
Microsoft will also use artificial intelligence to filter queries and use the same kind of signals that help Microsoft Defender filter out problematic websites, for example. These blocklists and classifiers will be used to “reduce the risk of abusive challenges being used,” Microsoft said.
Interestingly, Microsoft is also using additional technology to address distortions found in AI image generation. (Microsoft hasn’t clarified what this means, although anecdotally it seems that certain general challenges favor results with certain skin tones.)
“We take our commitment to responsible AI seriously,” Microsoft said. “To prevent the delivery of inappropriate results in the Designer and Image Creator applications, we are working together with our partner OpenAI, who developed DALL∙E 2, to take the necessary steps and will continue to develop our approach. We will regularly take the feedback we have and share it with OpenAI to improve the model and also use it for our own mitigation work.
Microsoft said the generation of its images will follow its content policy, which prevents images of child sexual abuse, involuntary intimate activities, suicides, terrorism, hate speech and more.
#Microsofts #art #generator #spark #imagination