Everyone Can Be an Artist with AI!

Yes, you too can be an artist with AI! Generating art with machine learning has taken off since Google's DeepDream program in 2015, and it has been getting better and better ever since. The DALL·E program by OpenAI took AI-generated art to the next level, especially with the new DALL·E 2. In this article I will introduce you to Google DeepDream and OpenAI’s DALL.E Program. I guarantee that you’ll be blown away by how great AI is becoming for generating art after reading this article.

Google DeepDream

Have you ever wondered how computers see our world? In 2015, Google revealed a trippy program called DeepDream, a computer vision program that visualizes the patterns learned by neural networks of computers. Like when a child watches clouds and tries to interpret random shapes, DeepDream over-interprets and enhances the patterns it sees in an image. As a result, DeepDream algorithms can produce dream-like portraits based on what the computer neural network "wanted" to see in an image as if it were dreaming. You can think of DeepDream as an AI filter that the computer applies to your images through what they see in their dreams.

In 2016, Google hosted an exhibition of art generated using DeepDream algorithms, together with Gray Area Foundation for the Arts in San Francisco, a non-profit organization that regularly hosts exhibitions focusing on interactive multimedia and technology. Some of the paintings were sold at a price of thousands of dollars. You can check out some of the work created using DeepDream here.

You can try how DeepDream works by using Deep Dream Generator, a platform built using Google DeepDream source code that allows you to input photos into the program and transform your photos through DeepDream algorithm. Google released the source code for DeepDream for free. If you know how to program, you can find their source code and play with it too. The image below is what Google DeepDream created for me after I input a photo that I took at the Lake Tahoe Emerald Bay. Left is the original image and right is image created using DeepDream algorithm. Looks cool, doesn't it? You can also check images other people created on the Deep Dream Generator website.

DALL.E by OpenAI

DALL·E completely took AI-generated art to the next new level. For Google DeepDream you need to input an image and the DeepDream algorithm will transform your picture for you. DALL·E creates realistic images from textual descriptions. So, you simply need to type a textual description in natural language and DALL·E algorithm will create a realistic image for you.

The DALL·E program was launched by OpenAI, an AI research and deployment company that was founded in 2015 by Elon Musk, Sam Altman, Greg Brockman, Ilya Sutskever, Wojciech Zaremba and John Schulman, who collectively pledged to invest over $1 billion into the venture when they set it up. The mission of OpenAI is to promote and develop friendly AI that benefits all of humanity. In 2019, Microsoft invested $1 billion in OpenAI. DALL·E 1 was revealed in January 2021. In April 2022, OpenAI introduced DALL·E 2, an updated system that generates more realistic and accurate images with 4x greater resolution compared to DALL·E 1.

If you enter a description “a painting of a fox sitting in a field at sunrise in the style of Claude Monet”, below are what DALL·E 1 and DALL·E 2 will create for you. Left is DALL·E 1 and Right is DALL·E 2. You can see how much more realistic and accurate images are created by DALL·E 2 compared to DALL·E 1. It's a huge improvement!

You should definitely watch the video below to see how magical DALL·E 2 is and how awesome the images DALL·E 2 created are. DALL·E 2 will definitely blow your mind!

If you would like to try how DALL·E 2 works, you should sign up here to be on their waitlist. Currently DALL·E 2 is not available to public, but it will be made available to a limited number of users who have signed up for their waitlist. I signed up when I first saw the DALL·E 2 news because it's so cool and I can’t wait to try it myself!