Three recent AI systems that generate remarkable art, given text descriptions. (None are available for use by the general public yet, but you can see examples of what they can do, and you can sign up for the waiting list for two of them.)
DALL-E 2 (made by OpenAI)
If you ask it to generate, say, “a painting of a fox sitting in a field at sunrise in the style of Claude Monet,” or “A bowl of soup that looks like a monster knitted out of wool,” it generates multiple images that do a remarkably good job of matching the text.
Their announcement post includes several examples and an explanatory 3-minute video.
That page also has a link you can follow to join their waiting list.
Imagen (made by Google)
Imagen does some of the same kinds of things as DALL-E. Their info page provides some examples, such as:
- A brain riding a rocketship heading towards the moon.
- A cute corgi lives in a house made out of sushi.
- A transparent sculpture of a duck made out of glass. The sculpture is in front of a painting of a landscape.
(But Imagen appears to be focused on creating photorealistic images; they don’t mention anything about the kinds of painted-in-the-style-of niftiness that DALL-E 2 can do.)
Midjourney (made by … Midjourney?)
Midjourney, “an independent research lab,” also does similar kinds of things. Some examples:
- fiery dragon as an architectural sketch
- Mandelbrot Fractal Sketches by Leonardo da Vinci
- Kandinsky music score
You can apply to join their beta.