Welcome to the future!!!

August 2022

Imagine if you could literally type any text at all and a program could generate it into an image? Imagine no more, it does exist! And there isn't just one of these. There are a number of projects that you can access right now. Here are a couple to try...

• Craiyon
• NightCafe

These are good, however, there is one A.I. project that is above them all. DALL.E 2 is by far the most powerful of these generators. I stumbled upon this amazing technology while I was surfing through YouTube about a week ago and proptly signed up for access (it is not open to the public yet). About a week later, I was pleasantly surprised to receive the invite to joinup. I initially received 50 free credits for the first month and 15 free credits each month after that. (You also have the option to buy more credits).

My experience so far

When you log in you are asked to type a detailed description and then press 'Generate'.

You can also choose the "Surprise me" and it will generate a phrase that you can either choose to "Generate" or modify. You can also upload an image to edit. Each time you generate an image you get 4 versions of that image. I saw that some of the early access people had 10 images per phrase, but to allow more people on the platform and spread the load of the high computing power required to generate images they have since reduced it to 4 images (at least for me). Hey, 4 is still awesome for me!

I have some examples below of the images I created with DALL.E 2. It is important to remember these are not copied from any other image. They are all origional. Even life like images are computer generated and do not exist anywhere else.

Without further ado, here are my initial efforts with help from my consultation team (the kids!):

“A donkey eating ice cream while riding a baboon in the australian outback”

(Our middle child helped me with this one):

hmmm... This was my very first effort. Not sure what to think, but not bad for a completely original computer generated image... but DALL.E did mix up the animals. I wanted a donkey riding a baboon, DALL.E swapped them around, which makes sense as that is a far more logical combination I suppose.

“aliens invading earth from the dark side of the moon”

“a cat submarine chimera, digital art”

This time I clicked the "Suprise me" button and it generated the text above. I then pressed "Generate" and created these.

Wow! DALL.E seems to produce really good digital art.

“The Mona Lisa with a moustache holding a fuzzy bear. oil painting”

Our eldest child came up with this masterpiece!

"A portrait of a black cat dressed in a tuxedo smoking a cigar"

We have a cat called Tux. He is black with a white patch on his chest that looks like a tuxedo. So this image was in his honour. The final image is definitely the best.

"A rabbit driving a submarine, digital art"

Our youngest child was inspired by the cat submarine chimera - this was her effort.

"A cat dressed as the king of Italy eating pizza"

Final thoughts

Is this the end of artists and illustrators? I dont think so, at least, not yet. What I think I can safely say is, this is cutting edge technology that will open up many possibilities that we cannot even imagine yet.

DALL.E 2 is definetly worth checking out. You can head to https://openai.com/ or join up to the waitlist here...

It might also be worth watching the following video from Marques Brownlee. He explains the whole technology very well.