The Times They Are a-Changin
Photo by Pavel Danilyuk: on Pexels
If you haven’t noticed, AI-generated images are taking the internet by storm. And let me tell you one thing. This is just the beginning. AI-generated visuals are here to stay. And it will sure as hell change the order of things.
For one it will probably change the way that artists work. Make it easier to conceptualize ideas while also making it more difficult to compete with the sheer amount of images these tools can create.
And the way to operate these tools is an art in itself. Thus a new profession soon may come into existence. AI whisperer. I came across this term while researching copyright issues regarding AI-generated images.
Mindblowing capabilities of AI
With huge conglomerates like Google, Meta (former Facebook), Microsoft, and companies like OpenAI and MidJourney, each working on their own versions of image-generating AI the whole space is progressing rapidly.
But although a lot of progress is yet to come. We don’t have to wait to see mindblowing results achieved using these powerful tools. And you can see a lot of those all over the internet. But here are some of my favorite examples:
1. A story on Medium by Jim Clyde Monge that went viral and made the author over $700
Showcases what DALL-E2 is capable of in a rather creative way. By using book descriptions of Harry Potter characters as prompts to generate the images.
You can join the waitlist to try DALL-E yourself here. I recently joined the waitlist myself and from what I’ve heard it might take around 4 weeks to get access. But it may vary.
2. A video by Two minute papers on YouTube
Showcases how DALL-E tackles outpainting. Enabling us to see the AI’s interpretation of what's going on around famous artworks like the Mona Lisa and more.
3. The public gallery of MidJourney AI
Here you can see some of the best images generated by MidJourney.AI. And by hovering the mouse over the image you can also see the prompts (text inputs used to generate the images).
You can also try MidJourney yourself. All you need to do so is Discord. Go to their homepage, press join the Beta and you will get invited to their discord server. Once you get there you should join one of the “newbies” channels. Then write a command /image and go crazy with your prompts. I tried it, was amazed, and highly recommend giving it a shot.
The thing about how these AI-powered image generators work is that each time an image is generated a graphics processor (GPU) somewhere needs to work its magic. And GPU’s consume electricity when performing these operations. This and the cost of GPU’s themselves and containing the whole operation create costs for the providers of these services.
And it currently looks that the pricing plans for these services will probably be based on how many images you generate. Here is where the AI-whisperer comes in. Someone who has the skill set to create the best images, with the least amount of tries. Someone who is able to create prompts that work the best for the AI tool in question.
Skills this person should have would include an eye for quality designs, technical knowledge of image parameters (Aspect ratios, resolution, etc), a way with words and a certain understanding of how the AI operates and where it draws, for a lack of a better word, inspiration for image generation.
AI image generation is here, and a new set of skills might be necessary to achieve optimal results using these mindblowing tools. Maybe this is the dawn of a new profession — AI whisperer. A fusion of a designer, a copywriter, and a programmer.
And although we are talking about image generation here, that’s just the beginning of AI entering the realm of creatives. Learn about these tools, adapt, and use them to thrive in this brave new world we are living in.
If you want to read more about AI, check out this list here. I will be adding new stuff to it over time.