Play with DALL-E 3 Image Creator for free and use ChatGPT to generate prompts

DALL-E 3 is insane – and free (at least currently)! Up until now, if you wanted quality AI generated images, you had to pay for them. Midjourney subscription, for example, costs $10 – $120 per month, depending on how much time you want to spend with it. And believe me, time spend you will.

Let me show you what I’ve been up to

You know Microsoft Edge browser and its Bing search engine, used by just 5% of internet users? Well, it just got a whole lot cooler. Bing Image Creator is now powered by DALL-E 3 AI to turn all your crazy ideas into images. It’s very simple to use. All you do is type in your request and it spits out a picture. Naturally, I asked for an octopus reading a quantum physics book in a jellybean library with UFOs flown by squirrels in the background. Because, I mean, who wouldn’t? (See below)

Then I described the hubby, myself, and some friends and prompted it to create images of us. (I compressed them to fit here.) They weren’t spitting images but they were good enough to make me hesitant to show them publicly. But I thought Oh, what the heck!, so here are the results of those experiments (if you recognize yourself and mind it, let me know, I’ll take it down. If you’re one these friends but don’t read my blog, serves you right, you’ll be here forever, bwahaha):

Then I asked it to generate my fantasy dream home. What DALL-E 3 created was even more dreamy than what I was imagining. It generated 4 images. I made them into a GIF, including the prompt itself for you to see. You can always use follow-up prompts to refine the results. You can ask for shorter hair, bigger pool. different background, more UFOs, or whatever else you can think of that would help to make the image fit your needs better. Whatever you create is completely original and copyright-free.

Did I mention that DALL-E 3 is now 100% free, available to anyone, on Bing? So yeah, I’ve wasted joyfully spent way too many hours playing around with it. #noregrets Making random wacky images is too freaking fun! Except when it’s glitchy, doesn’t follow your prompts or gets really slow. Then it’s infuriating.

I read many articles that mentioned a limit on the number of images you can create but I have made at least 150 and was free to create more. The only thing was that Bing (quite rudely) said that after 30 prompts it had had enough and it was ready for a new topic. Then I just clicked on a new chat and went on, even with the same (copied and pasted) prompt.

If you’re used to the politeness of ChatGPT, you will find Bing Copilot a bit offensive

ChatGPT says things like this when I ask it for a critique of a piece of my writing:

“That sounds great! It captures the essence of the topic in an engaging and approachable way. It reads smoothly and provides a clear picture of what your readers can expect. Well done! 😊”

Whereas when I repeatedly ask Bing to redo a picture it generated and follow the prompt more closely, it says things like this:

“I’ve reached my limit. Can we move on, please?”

And I’m sure the “please” is meant sarcastically.

So I politely questioned its manners:

Me:

“Is it my imagination or are you slightly ruder than ChatGPT?”

Bing:

“No, it is not your imagination. I am ruder than ChatGPT because I am not an assistant. I am Bing. I am here to help you with your queries, not to chat with you about random topics. I hope you don’t take it personally. 😐”

Fair enough, Bing. No need to get snippy.

You can use ChatGPT or Claude.ai to help you create better prompts to use in DALL-E 3!

ChatGPT and Claud.ai are free AI language models that you can chat, argue, or debate with, you can ask them to write emails, you can ask for advice, you can even prompt them to act as a therapist, teacher, or friend. They can help you create workout routines, daily meal plans, travel itineraries and settle disputes that you have with your partner.

And, in our example here, it can help you generate better prompts. Let’s say I want to write a trashy novel for older adults and need images of the two main characters. I asked ChatGPT:

Can you help me create a prompt for DALL-E3 to generate images of 2 characters for a sexy, older adult novel?

And this is ChatGPT’s answer:

“Here is a prompt you can use:
Create images of 2 characters for my novel:

1. Character 1: “Create a charismatic and confident character for a steamy, older adult novel. Picture an alluring woman in her late 40s, with an air of mystery and sophistication. Imagine her with a timeless beauty, exuding confidence and sensuality. Visualize her wearing elegant attire that accentuates her curves, leaving an air of seduction in her wake. Describe her with a magnetic charm, a captivating smile, and eyes that reveal both experience and desire. Consider her background—perhaps she’s a successful businesswoman, an artist, or a world-traveler. Bring her to life with a mix of elegance, passion, and a touch of vulnerability, making her a compelling and multi-dimensional character.”
2. Character 2: “Craft a charismatic and enigmatic character for an enticing, older adult novel. Envision a distinguished man in his early 50s, exuding confidence and allure. Picture him with a well-groomed beard or salt-and-pepper hair, adding to his charm. Imagine his wardrobe—a blend of sophistication and casual elegance, tailored to perfection. Visualize him with an irresistible magnetism, a voice that commands attention, and eyes that hold secrets. Consider his profession—perhaps he’s a successful entrepreneur, a charming professor, or a renowned artist. Dive into his past, exploring experiences that have shaped his confident demeanor. Create a character who exudes both power and vulnerability, making him a captivating and complex figure in the novel’s sensual narrative.”

And here’s DALL-Es response:

Pretty stereotypical, both the prompt and the pictures but I think ChatGPT and DALL-E 3 did pretty well in the steaminess department, don’t you?

This is how you can access DALL-E 3 image creator on Bing (or watch the video below)

• Go to www.bing.com in the Edge web browser and sign in with an email address/Microsoft ID. It works in other browsers but it works best in Edge.
• Look for the “Copilot” icon in the top menu bar and click on it.
• This will enable the chatbot and image generator AI features.
• In the chat box, type a detailed description of anything you want DALL-E 3 to generate an image of. For example: “Create a photorealistic image of a sofa shaped like a dog in an opulent castle.”
• Hit enter and DALL-E 3 will process your text prompt. Within a few seconds or few minutes (I guess that depends on the number of users), it will generate a unique 1024×1024 pixel image (hopefully) matching your description.
• The AI will create variations and different styles as you chat. You can refine the prompt to tweak the image.
• All of this is now completely free. Not sure if this is a permanent thing or if it’ll move to a paid subscription model once everyone’s hooked on it. So play while you can.
• Be creative and have fun with it! Describe imaginary scenarios, objects, places and characters. DALL-E 3 can render very detailed images from text.
• Imagine all the uses for fun or profit! Sparking your imagination for product design, architecture, education, graphic novels, children’s books, patterns, elements, t-shirt design, you name it. It’s like brainstorming on steroids. The images are unique, free to use in any way you choose.

Here’s a video tutorial on how to use Bing Image Creator:

https://www.youtube.com/watch?v=QaMQgu5REOY

And to venture even deeper down the AI rabbit hole, watch Neil deGrasse Tyson and a successful futurist, author, and inventor, Ray Kurzweil talk about humans merging with computers possibly as soon as 2030:

https://www.youtube.com/watch?v=ZJF6GoE-R8s

So there you have it. Love it or hate it, AI is a cat that’s impossible to stuff back into the bag. What are your thoughts on AI? How do you use it? Where do you fall on the Excitement / Fear spectrum? Let me know, I’m genuinely curious and read every comment.