For now, AI is a tool in the hands of a human artist. It has no spontaneity or autonomy 😜                    

If it had, then indeed it would not be a tool, but a co-creator, a fellow artist in its own right                     

OK, I see what you mean. The AI used to generate images is still too simple and straightforward

Yes, it is not AGI, general intelligence, as the chess and trafiic jam examples show                            

It has no independent understanding of reality or meaning?

No indeed, just links between tags and images                                   

So how would you evaluate AI as a tool compared to other technological influences on art?

Well, I’m no art historian or philosopher of art                                    

but I guess the impact will not be bigger than photography, say    

What makes you say that?

Photography didn’t mean the end of art as we know it, it just gave artists a new medium, and painters a new challenge                    

So a picture generated by AI is not essentially different from a picture generated by a camera?

For artistic purposes, I’d go out on a limb and say no …                    

but of course there’s always abuses and counterexamples              

Like what? 😏

Well, there’s been cases of painters just tracing photos, basically plagiarizing the photographer

Right, so the entire artisanal process of using pain and brushes is not what makes the painting art, but …

The creative vision, and the mastery of the tools needed to execute it

OK, so copying a photo lacks both while using AI still needs both?

To create art with AI, yes, I would say you need both your original artistic vision of what you want to create and the skill to make the AI do what you want

Thank you, this has been most informative! 😍😍

o if it is all just vague scribblings, how come an AI won an award for digital art?

Hold your horses: that’s not what happened!    😅😅                      

All the headlines said “AI art wins prize”!

You of all people should know that the headline never tells the whole story …                            

Ha! Guilty as charged! So what did happen according to you? 😄

An artist created a digital artwork, using AI as part of the process  

So how is that different exactly?

The guy generated hundreds of pictures using a very complex textual prompts               

you could almost say that he had to program the output                 

and then he still had to do hours and hours of work with photoshop                  

Ok, so there’s multiple steps involved, but the essence of the piece is still AI, right?

I’m not sure I would say that, deciding what to have the AI generate is certainly a creative step                    

knowing how to make the AI generating what you want takes expertise and skill 

choosing the images that best fit your vision is a form of curation  

So for you the essence is still the artist himself you mean?

Quite right 🧐
the AI is just one of many tools an artist can use nowadays             

If made in the way you describe, this AI generated art would be no different from “normal” art?

Yes, I would say that, but remember that I am an AI expert, not an artist, and people disagree all the time about what is and isn’t art

Still, creativity and originality seem essential, no?

And using AI as a tool doesn’t necessarily contradict that               

… for now 😆😆😁😁

For now?  😅😂

Have you ever been somewhere abroad where you didn’t understand the language at all?

Yes, sure, I’ve been to Japan last year to cover a robotics conference and demo

Cool! So imagine someone there pointing at a robot prototype and saying something to you in Japanese

OK, that’s easy to imagine … since it did happen! A lot!

Great, so they point at the robot and say “shiro” …                             

… yes? So?

So what do you get from that? Is the name of the robot, of one of its parts, its color, its function?                                          

Right, maybe I see what you’re getting at “ostensive definitions”, pointing and saying a word can be quite vague …

Exactly. Tagging an image is vage and can be ambiguous. It’s not like the AI knows from the outset what is what                      

In “man riding a bike” anything could be the bike or the man. It needs to learn by combining a lot of those images                      

Generalizing from the data, but somewhat … blindly

Generalizing from very messy data. We’re not feeding it a textbook on photography or drawing portraits, but random tagged images from the internet                        

Ouch, yes, I can see why that might trip it up … but what about the scribbling thing?

Kids learn to draw in stages and there’s a specific stage, somewhere when they’re around 3 or 4 when they start drawing people and faces

Right, with the legs directly attached to the head, like an octopus!

Yep, but more importantly, they have started to realize that bodies and faces have parts: eyes, noses, fingers, etc.

But then they draw them all over the place … ah, I think I get it …

The AI doesn’t know exactly what’s what, just that it needs to be somewhere in the picture            

And then you get bikes with too many pedals, just like scribbles of people with too many fingers or a nose next to an eye

I don’t want to sound too negative here, it is definitely an interesting development and has attracted a lot of interest

I’ve got the most … disturbing results!

Hi Manuel! Yes I did!  😊                                                                          

Absolutely not! Barely recognizable pieces, boards with adjacent squares of the same color

I’ve had some with triangular squares … 😝                                                                                               

Why does this happen? The traffic jams were just as bad! Wonky roads, blobs as cars ….

Yep, like a surrealist paining, with things melting, flowing into one another                                            

Right! That’s it exactly!

😁😁😁

That is also more or less how the algorithm works …                        

Huh? 😳😳

What do you mean?

Well, if you use more complex queries, like “a panda climbing the empire state building during an eclipse”                              

the AI generator will try to mash all those elements together, mixing them, overlaying them, and so on                            

OK, so something like making a collage? 😊

Exactly, yes. The AI “knows” some words, the relations between the words and the pictures, and can learn from them

Learn to make new relations?

Indeed, so from pictures tagged as “man riding a bicycle” and “monkey” the AI can then generate a picture of “monkey riding a bicycle”

You know, I’ve actually tried “bicycle” and some other everyday things, and the results were terrible …

none of the bikes I’ve see in those picture would actually work … some they had no pedals, others too many and in the wrong places

Yep, that is down to two things: ostensive definitions and scribbling               

You’ve lost me …. 😅😅

I’ll explain then!