DALL·E 2 - the TALENTED robot artist
BRINGING ALL OF YOUR CREATIVE IDEAS TO LIFE IN SECONDS
Ahoooooj everyone! I have some exciting news today! I got access to OpenAI DALL·E 2!
Why is this making me so insanely happy? DALL·E 2 is a new AI system that generates images from text prompts. That means whatever you can describe in a natural language, DALL·E 2 will create.
It can combine concepts, art styles, even effects of different lenses! *mind blown*
There's a million-person wait list at the moment so I feel very honored that I got this opportunity and I can’t express how grateful I am that OpenAI reached out to me!
On Wednesday I was a part of an onboarding Zoom call with Natalie and Andrew - two lovely and fun people from OpenAI, and they explained the rules for using DALL·E 2 so it’s safe, and the creations don’t spread misinformation or deepfakes and whatnot and answered some Q&A questions.
My first prompt was quite long but I had lot of fun with if - you can create variations of the image DALL·E 2 generated, or adjust parts of the image you didn’t like.
There are so many possibilities that this can be used for, from quick story boards, backgrounds for projector art, song covers, photoshop matte paintings, AR filter ideas, 3D objects and so much more.
I don’t think AI art will put human artists out of business - I am an artist myself so why would I dig myself a grave? Artificial Intelligence is for the 21st century what electricity was for the 20th. I believe this will expand the horizons in the most intriquing ways.
I wanted to see how DALL·E 2 will tackle the task of creating clothing designs, especially with adding some more specific things as regional and traditional clothing, that’s unique and from my country of Czech Republic. I added a little futuristic twist to it by requesting neon coloring and the results were pretty amazing!
OpenAI hope is that DALL·E 2 will empower people to express themselves creatively. DALL·E 2 also helps us understand how advanced AI systems see and understand our world, which is critical to our mission of creating AI that benefits humanity.
“DALL·E 2 has learned the relationship between images and the text used to describe them. It uses a process called “diffusion,” which starts with a pattern of random dots and gradually alters that pattern towards an image when it recognizes specific aspects of that image.”
Who doesn’t love cute puppies and cute animals in general? Everyone! Here are some results from my creations including dog and cat prompts. And space. And robots… I mean you know me right? :D
Alright that’s it from my very first attempts and generations with DALL·E 2, I am very impressed. You have to be specific with the prompts and learn how to use them in a way it gives you the results you desire. I have a text document where I write down all of my prompts, and keep adding and tweaking different ones until I get the desired results. Keeping track of your prompts is a great way how to keep track of what works and what doesn’t!
Let me know in the comments what would you like to see DALL·E 2 create next or what are your thoughts about this technology!
