Is Copilot the best AI companion out there? Help us find out by answering a couple of quick questions!
— Bing (@bing) March 21, 2023
Starting today, some who have access to the preview of the new Bing will also be able to generate images using natural language. In its announcement, Microsoft said the arrival of Bing Image Creator to the ChatGPT-powered Bing and Edge, allowing it to accept instructions to create AI images using OpenAI’s DALL?E model.
“We’re excited to announce we are bringing Bing Image Creator, new AI-powered visual Stories and updated Knowledge Cards to the new Bing and Edge preview,” said Yusuf Mehdi, Corporate Vice President & Consumer Chief Marketing Officer, who said it will be available both on the mobile and desktop version of Bing. “Powered by an advanced version of the DALL?E model from our partners at OpenAI, Bing Image Creator allows you to create an image simply by using your own words to describe the picture you want to see. Now you can generate both written and visual content in one place, from within chat.”
According to Mehdi, who called the new feature a “creative copilot,” it is now being rolled out to those with preview access to the new Bing. On a side note, this release will come in phase, which explains why many can’t still access it, including our team. Meanwhile, to those who already have it, it will be accessible in the Creative mode of the chatbot (but Microsoft said it has plans to bring it to all modes and in the new Bing button in chat mode in the preview version of Edge soon). Those who are still waiting can go to bing.com/create instead.
As Microsoft noted, the generator can accept specific instructions, allowing users to specify even the art style and other details of the images they want Bing to create.
Interestingly, this rollout also covers Edge, meaning users can soon generate images using the sidebar of Microsoft’s browser. Here, Edge will offer a Bing Image Creator icon to access the feature.
On the other hand, while the news sounds big and compelling, Microsoft was swift to explain its principles for promoting and integrating AI into its products and services, including the steps it made to ensure the safe use of Bing Image Creator. This is expected, nonetheless, given the past issues and lawsuits regarding AI image generators.
“At Microsoft, our teams are guided by our Responsible AI principles and the Responsible AI Standard to help them develop and deploy AI systems responsibly,” explained Mehdi. “To curb the potential misuse of Image creator, we are working together with our partner OpenAI, who developed DALL?E, to deliver an experience that encourages responsible use of Image Creator. We have ensured OpenAI’s safeguards, plus additional protections, have been incorporated into Image Creator.”
In particular, Mehdi said there are some restrictions on the use of Bing Image Creator, just like Sydney itself. This includes blocking entries instructing Bing to create harmful or unsafe images. Also, as shown by Microsoft, the images that will be generated will come with watermarks, which is probably aimed at discouraging the improper use of images. However, given the continuous attempts of different users to break Bing (which gave us the malfunctioning Sydney), time will tell how effective these restrictions are.