A variety of illustrations made through the artificial intelligence Midjourney , which shares a name with the research lab that developed it, were shared on a popular Japanese comment forum . This artificial intelligence creates images from textual descriptions, similar to OpenAI 's DALL -E , and is currently in open beta. The Midjourney team is led by David Holz , co-founder of Leap Motion .
According to Twitter user “ 852話 (@8co28) “, the team added some updates recently, and someone decided to try the option of “2D Kawaii” as one of the terms ended up generating the following illustrations:
The tool is only available through the official Discord server and works with a command " /imagine ", followed by a phrase that describes what the user wants to achieve. However, the test is limited to twenty-five uses per person (something that can be avoided by simply creating a new Discord account), and the base phrase must be written entirely in English. Of course, all of this just makes the artificial intelligence train itself and deliver ever more impressive results. Is this the democratization of art? In other words, does this mean that now everyone will have the same opportunities in the art world?
It turns out that the first video game that uses illustrations generated through this artificial intelligence, entitled " Kabeyume (壁夢) ", launched on August 12 through Nicovideo Games in Japan, has already been released.
Source: Yaraon!
