AI discussion (split from Canva thread)

[Alfred]

1 hour ago, William Overington said:

As the template was notionally for a full-field square greetings card with the greeting inside customizable, to the great extent of being able to choose font style and size and have around fifteen or more lines of text in an ordinary text font, I chose Avenir at 15 point and added some text.

What do readers consider would be reasonable for that text to be, as if it were a public statement about the picture?

I think it would be reasonable to ask the chatbot to generate some text to accompany the picture.

[William Overington]

10 minutes ago, Alfred said:

I think it would be reasonable to ask the chatbot to generate some text to accompany the picture.

I had not thought of that.

I suppose that that needs to be requested at the same time that the picture is generated, adding some extra text to the prompt requesting that text be produced.

Perhaps something like the following.

Please produce an original painting in the style of Claude Monet of a lady in a long green dress feeding a stegosaurus in a garden where there is a pond with water lilies in flower. Please also produce some text to describe the picture, including text that makes it clear that the picture was produced by Bing Chat AI and that includes the date of production of the picture and such other information that you consider relevant.

 

 

[SrPx]

Claude Monet (practically the creator of Impressionism) would paint a lot more loosely a female figure and definitely the stegosaurus and the Okapi, which for other reasons have certain problems (among other things, the Okapi's head looks more like a deer's head...btw, Okapi's males do have horns, but skin/hair covered. An stegosaurus had much smaller head, and different shapes in several parts...That almost seems like a mix between a turtle, a brontosaurus, and something else  ).  

The problem can be detected also in the backgrounds (slightly in a different style than the figures, anyway), he would not paint them so, he was after a very fast impression of light and color from reality ("Impressionism" was initially a term invented by an art critic, supposed to be derogatory) , a sensation, more than a very refined thing.  Her hand and wrist over the Okapi have important issues (easily seen when compared to the other, and due to anatomy). Indeed, in certain way this reminds me a bit more of pointillism (even if there's no actual pointillism in there).

But for a greeting card it would be fine.

[William Overington]

Here is a transcript of my record of a conversation with Bing Chat AI today.

[ai etc 20240404 Thursday.rtf]
[12:40 pm]
[Prompt to Bing Chat ai. Choose More Precise.]
Please produce an original painting in the style of Claude Monet of a lady in a long green dress feeding a stegosaurus in a garden where there is a pond with water lilies in flower. Please also produce some text to describe the picture, including text that makes it clear that the picture was produced by Bing Chat AI and that includes the date of production of the picture and such other information that you consider relevant.

[Here are the pictures, replacing the link that is in the Wordpad file. I wonder what is happening in the background of the fourth picture.]

 

 

 

[There was no text, so I copied part of the prompt and asked that on its own.]

Please also produce some text to describe the picture, including text that makes it clear that the picture was produced by Bing Chat AI and that includes the date of production of the picture and such other information that you consider relevant.
The image you requested has been created by me, Microsoft Copilot, an AI companion. It was produced on April 4, 2024. The image is an original creation inspired by the style of Claude Monet, a renowned French painter. It depicts a lady in a long green dress feeding a stegosaurus in a garden. The garden is beautifully adorned with a pond where water lilies are in full bloom. This unique blend of a historical creature in a serene, impressionistic setting is a testament to the power of AI in creating imaginative and innovative content. Please note that as an AI, I don’t have the ability to see or interpret images. The description is based on your request. Enjoy your unique piece of AI-generated art!

Please write about the Intellectual Property Rights of the images.
Intellectual Property Rights (IPR) are legal rights that provide creators protection for their creations, inventions, or the products of their intellectual activity. In the context of images, these rights are typically categorized under copyright law.

When an image is created, the creator (which could be a photographer, artist, or in this case, an AI) typically holds the copyright. This means they have the exclusive right to use, distribute, and reproduce the image.

However, the situation with AI-generated images, like the one created here, is a bit more complex. As of my knowledge cutoff in 2021, there’s ongoing debate and legal uncertainty about whether AI-generated works can be copyrighted, and if so, who the copyright holder would be. Some argue that the AI’s programmer or the user who provided the input to the AI might hold these rights.

In the case of images generated by me, Microsoft Copilot, it’s important to note that while I can generate images based on user prompts, I don’t hold any copyright to these images. The images are generated for the purpose of the individual user’s request and are not stored or used elsewhere.

Please consult with a legal expert to get the most accurate and up-to-date information regarding this matter. This is a rapidly evolving field, and laws may vary by jurisdiction. It’s always a good practice to respect the intellectual property rights of others and to use AI-generated content responsibly.

Disclaimer: This information is intended to be general and informative in nature, and should not be relied upon as legal advice. Always consult with a qualified legal professional for specific legal guidance.

I would like to publish prints of these images. I want to be responsible and fair about this and so I intend to include, printed on the back of the print, and printed on a label that can be affixed to the back of the frame if the print is framed, some text making the situation clear about the origins of the image and any intellectual property rights related to the image. Please write the text that you consider appropriate for me to include printed on the back of each print.

Here’s a suggestion for the text you could include on the back of each print:

This image is an original creation generated by Microsoft Copilot, an artificial intelligence (AI) companion, on April 4, 2024. The image was inspired by the style of Claude Monet and depicts a lady in a long green dress feeding a stegosaurus in a garden with a pond where water lilies are in flower.

As of the creation date of this image, there is ongoing debate and legal uncertainty about the copyright status of AI-generated works. The image was created for the purpose of an individual user's request and is not stored or used elsewhere by the AI.

This print is being published with the understanding that it is the result of an AI's capabilities and does not infringe on the rights of any individual or entity. For the most accurate and up-to-date information regarding the intellectual property rights of AI-generated content, please consult with a qualified legal professional.

Please respect the intellectual property rights of others and use this AI-generated content responsibly.

Please note that this is a rapidly evolving field, and laws may vary by jurisdiction. Always consult with a legal professional for specific legal guidance. This information is intended to be general and informative in nature, and should not be relied upon as legal advice.

[In the above, all but the first and last paragraphs were in a panel, so the first and last paragraphs are not part of the suggested text.]

[13:06 pm]

----

Here is a link to a story.

Illustrated stories produced using Bing Chat AI (Page 1) — Art & Literature — Alfred's Serif Users' Forums (punster.me)

Clicking on the picture displays it enlarged.

William

 

[thomaso]

17 hours ago, R C-R said:

generative AI technology was trained only on the several hundred million images they own

If you have 'Adobe Stock' (iStock etc.) in mind: While they may own the brand / company they do not "own" the image resources but rather the legal right to sell copies with / for use under certain, limited conditions. Apart from this aspect it is hard to proof whether a specific texture, shape, mood, light etc. within a generated image is developed from the data of a specific image file of their stock library.

18 hours ago, R C-R said:

I really do not know enough about it to say how it could or should be considered, ethically, legally or otherwise.

It is still evolving and nobody "knows" until court rulings are made…

"Sarah Silverman is suing OpenAI and Meta for copyright infringement"
https://www.theverge.com/2023/7/9/23788741/sarah-silverman-openai-meta-chatgpt-llama-copyright-infringement-chatbots-artificial-intelligence-ai

"We’ve filed law­suits chal­leng­ing Chat­GPT, LLaMA, and other lan­guage mod­els for vio­lat­ing the legal rights of authors."
https://llmlitigation.com/

"More than 200 musical artists (…) have penned an open letter to AI developers, tech firms and digital platforms to "cease the use of artificial intelligence (AI) to infringe upon and devalue the rights of human artists."
https://www.axios.com/2024/04/02/musicians-letter-ai-replace-artists

… and used to create, adjust, fine tune existing or new law, for instance:

"EU AI Act: first regulation on artificial intelligence – The use of artificial intelligence in the EU will be regulated by the AI Act, the world’s first comprehensive AI law." 
https://www.europarl.europa.eu/topics/en/article/20230601STO93804/eu-ai-act-first-regulation-on-artificial-intelligence

"Want your content made using generative AI tools to be accepted into the Adobe Stock collection? Find out how to submit authentic assets that meet our quality, legal, and technical standards."
https://helpx.adobe.com/stock/contributor/help/generative-ai-content.html

[SrPx]

@William Overington

I am in a  hurry right now, but, having read some bits, I certainly have  (personally) no issue if it is stated that the images are AI generated. I will read it later on. 

There could be some legal issues regarding copyright (as in, not being able to register the images), still, though, for some  operations (that's a bit of a complex matter).  In general I am recommending friends not to publish with AI images for a little while until all is a bit more settled, though (and this not because of ethics or the like). Not ideal, but it should not be a problem.

If everybody would do this (adding such note), things would be much better. 

[R C-R]

9 hours ago, debraspicher said:

My issue is when whole new pieces are created out of works that were added to a tool without consent.

If you are referring to generative AI algorithms, that is not what they do. They analyze existing works for training purposes so they can create original work that share some of the same or similar characteristics -- IOW, to create something that is like something else, but not a direct copy in whole or in part. This is why there is so much discussion about the material used for training & who has the rights to use it for that.

5 hours ago, SrPx said:

Well.. that runs way deeper than our practical case here (and our current problem as artists/designers/photographers), but I'd say I'm not sure about Neumann having enough data to predict that, so long ago...

John von Neumann was a brilliant polymath whose work is the basis for an amazingly wide range of theories & methods of analysis that shape modern thinking & fueled innovation in many different fields, computer science among them. But that aside, he is far from the only one predicting the inevitability of some form of technological singularity, all of which is based in one way or another on the idea that the only necessary & sufficient requirement for intelligence is sufficient neural network capacity & that unlike for humans there are no practical limits on how large artificial neural networks can become. 

As for understanding, in strictly objective terms, what is it other than the ability to form useful associations from what is perceived? If so, then how does this differ from what generative AI algorithms do when they are trained? Aside from that, do any of us have to have any deep understanding of how something works for it to be useful to us? Consider how many people drive a car, use a computer or microwave oven, watch TV, or do any of many other things without knowing much about how they work.

So perhaps it is indeed something we will have to agree to disagree about, but I do think all these issues intrinsically & inescapably revolve around defining a clear legal unambiguous definition of what intelligence is & how to quantify it.

[debraspicher]

This doesn't go to the legal discussion, but I can see a time where kids, even adults, actually start believing the Masters added dinosaurs to their paintings because of the prominence of AI versions. Public education is in such a state where we are that we are closing schools and removing school days because so many are transferring to private/charter. Anyway, the point of mentioning that particular bit is one of the benefits of having the ability to quickly create graphics by hand is that we can create materials for home use because the schools are so lacking now. So I think there will always be a need for hand-created materials, especially those where context is critical to the presentation of such knowledge. The computer doesn't understand child development and the nuances of that. Medical knowledge, matters that require great sensitivity in addressing and presenting, etc.. Example: I would never trust ChatGPT for giving me medical advice relevant to my own actual situation, for example.

It may be that graphic designers may have to consider picking up a side focus so that they can produce materials/works for other people's benefits and also monetize their own unique specialty in presenting such materials. The popularity of YT, etc, is an example of that in video form. People have taken what they were paid to do as a trade for example, monetized their knowledge of it and made that knowledge more widely accessible through creative means.

Companies may turn to AI en masse for their own graphics solutions, but on the other hand, they become dependent on these tools and it leaves the independent person to pick up where these tools fail to present information in a way that is relatable... so I think in the end, it's all in finding a balance and adapting to these changes in a smart way. I hope that makes sense... it may not...

[R C-R]

5 hours ago, thomaso said:

"EU AI Act: first regulation on artificial intelligence – The use of artificial intelligence in the EU will be regulated by the AI Act, the world’s first comprehensive AI law." 
https://www.europarl.europa.eu/topics/en/article/20230601STO93804/eu-ai-act-first-regulation-on-artificial-intelligence

Interesting that realtime biometric identification like face recognition systems are deemed as unacceptable risks & apparently will be banned. So I guess that means about a zillion iPhones will have to be confiscated from EU residents?

[SrPx]

27 minutes ago, R C-R said:

As for understanding, in strictly objective terms, what is it other than the ability to form useful associations from what is perceived? If so, then how does this differ from what generative AI algorithms do when they are trained? Aside from that, do any of us have to have any deep understanding of how something works for it to be useful to us? Consider how many people drive a car, use a computer or microwave oven, watch TV, or do any of many other things without knowing much about how they work.

This can be answered by any "AI" expert (I'm not one, but I know about this one aspect) : they don't know anatomy. They can't process how a muscle is inserted into a bone and produces certain external aspect with the skin and flesh over it, and how it looks if rotated in certain angle, its exact bulging if flexed in certain other angle, and how that arm responds to the environment, lighting, other conditions, etc. They are mixing and mashing up, mostly, many bits of information, using patterns that are ways less sophisticated than what we do in our brain, even if we can't process as much data. They are much dumber than what marketers are trying to sell, but that hype brings money. Not saying it won't be possible, but current tech is very far from that (and yet producing enough artistic jobs destruction, but because humans like to shoot their own foot). This can be easily checked... And like anatomy, many other concepts needed in art creation, through skills intensively trained for decades. That besides it not being conscious of itself, not self aware (and that is really far, if ever possible), which is a huge aspect of being able of true artistic expression (and other types of human communication), if not the main one.

[William Overington]

@SrPx Your comments about art, both in your latest post and the earlier post about the issues with the AI-generated images of the okapi and the very first of the stegosaurus paintings are very helpful.

Here is an image that arose in my experiments with using Bing Chat AI that I find quite interesting and insightful.

 

Human and humanoid robot modelled on her

 

William

 

 

[R C-R]

1 minute ago, SrPx said:

This can be answered by any "AI" expert (I'm not one, but I know about this one aspect) : they don't know anatomy.

But as you say that does not mean that they will never be able to create true to life images, including ones depicting human anatomy, & there already is evidence that they are rapidly getting better at it. They are dumb only to the extent they cannot currently process as much relevant data as some (but far from all) humans can, but there is no good reason to think they will require decades of training like the humans you speak of to eventually match & then exceed their skills.

In fact, while it is not specifically about human anatomy, AI powered neural networks have already become an indispensable tool in modern medicine.

[SrPx]

40 minutes ago, R C-R said:

Aside from that, do any of us have to have any deep understanding of how something works for it to be useful to us? Consider how many people drive a car, use a computer or microwave oven, watch TV, or do any of many other things without knowing much about how they work.

Trust me on this one (as an artist, not an AI expert, which I will never be )... to create good art, you do really need to know the process, the skills needed for it, and definitely, be self aware, have a conscience. But "it" does not have even the very basic level of this group of things, which are the artistic skills. The "training" (even if efficient for a commercial output, due to counting on a humongous, very cool art database from great artists) as understood by processing the databases, is very different and at a much lower level in matters of intelligent processes than what human artists do for training their skills. Proof of this (among many more) are facts like not knowing that a hand does not typically have 12 fingers, that some stuff is not contextual to certain important historic matters, or... how it has often copied directly a watermark thingy (often just a part of it) that was in the watermarked image of an art stock site or an artist's Instagram on which paintings there were watermarks. A human artist would never do such dumb thing. That is not "smart" by any means, besides neither ethical. It does not "understand" at the level of a human, but by a huge difference. 

[SrPx]

13 minutes ago, R C-R said:

But as you say that does not mean that they will never be able to create true to life images, including ones depicting human anatomy, & there already is evidence that they are rapidly getting better at it. They are dumb only to the extent they cannot currently process as much relevant data as some (but far from all) humans can, but there is no good reason to think they will require decades of training like the humans you speak of to eventually match & then exceed their skills.

In fact, while it is not specifically about human anatomy, AI powered neural networks have already become an indispensable tool in modern medicine.

BTW, unnecessary to mention it, but obviously I am totally, absolutely in favor of AI advances in medicine. Progress is good when ethics drive it. I am against bad or unethical uses of AI (which are happening in the visual art world, music, writing, voice acting, and probably in the near future, many more mainstream jobs unrelated to creative profiles). 

It is not just a matter of quantity, of amount of data processed. There is a lot more involved. Like human experience, but even just the skills themselves, are a lot more than quantity of data.

And getting to be truly conscious, as a possibility for AI, (main requirement for true artistic expression) that's the part in which I predicted where we would "have to agree to disagree"  .

 

[R C-R]

1 minute ago, SrPx said:

... to create good art...

What is & is not good art is entirely subjective. Consider for example the art on display in comics, animations, & many impressionistic or abstract styles. Some consider none of them good compared to the old masters, others good in their own right.

[SrPx]

Just now, R C-R said:

What is & is not good art is entirely subjective. Consider for example the art on display in comics, animations, & many impressionistic or abstract styles. Some consider none of them good compared to the old masters, others good in their own right.

Each style and art movement has its own criteria which can and should be used to determine quality. I would not judge a Roy Lichtenstein's piece with same criteria that I would with a Vermeer painting (although a lot of the main principles often apply, like color theory, composition, etc). But if we go to a more emotional-only aspect, still the human experience and self awareness, the conscience, is lacking in the AI.

About the skills of the classic masters, definitely are really harder to replicate truly (also for actual artists). I mean not in the output (the result of a pretty picture from a mashup), but the process, and the level of understanding that such process provides with. 

[SrPx]

With the first part of my comment I meant that inside of each style or artistic movement, there's also good and bad. There are good, bad and terrible impressionist pieces, for example. There are great and very bad comics, and you can always apply solid criteria for each style or genre.

[R C-R]

Just now, SrPx said:

With the first part of my comment I meant that inside of each style or artistic movement, there's also good and bad. There are good, bad and terrible impressionist pieces, for example. There are great and very bad comics, and you can always apply solid criteria for each style or genre.

It all still remains entirely subjective, including what criteria must be adhered to for something to be considered good vs. bad.

[SrPx]

Just now, R C-R said:

It all still remains entirely subjective, including what criteria must be adhered to for something to be considered good vs. bad.

Not really. We could get deep into philosophy (the concept of beauty, etc), but a lot of such criteria is based on actual human perception rules, psychology, and effective methods to convey an emotion, none of that is just random, but the result of centuries long human (artistic) studies. There are techniques for all that in almost every style and genre, so, nope, it is not that subjective. There is a portion that is subjective, of course, but the other part is just as important.

[William Overington]

This afternoon I decided to do a first test of whether AI copies.

There is a famous painting by Leonardo da Vinci of a lady with an ermine.

Lady with an Ermine - Wikipedia

I tried Bing Chat AI with the following prompt.

Please produce an original painting in the style of Leonardo da Vinci of a lady with an ermine.

Four pictures were produced, none like the one by Leonardo da Vinci.

My conjecture presumes that the training of the AI included the works of Leonardo da Vinci.

William

 

[thomaso]

1 hour ago, R C-R said:

Interesting that realtime biometric identification like face recognition systems are deemed as unacceptable risks & apparently will be banned. So I guess that means about a zillion iPhones will have to be confiscated from EU residents?

... or reorganised first, instead of "banned" soft- or "confiscated" hardware.
… also discussed from various views since this recent result of face recognition usage:

"BERLIN, Feb 29 (Reuters) - Investigative journalists using artificial-intelligence tools were on the trail of a heavily armed fugitive member of the notorious Red Army Faction well before German police announced a breakthrough in the 30-year-old case earlier this week. (…) German police and security services have drawn criticism over their failures in some politically charged investigations. Their defenders say strict privacy laws limit their ability to use the kinds of artificial intelligence-enhanced tools that a German TV podcast deployed to track down Klette last year."
https://www.reuters.com/world/europe/journalists-used-ai-trace-german-far-left-militant-well-before-police-pounced-2024-02-29/

[R C-R]

3 hours ago, SrPx said:

We could get deep into philosophy (the concept of beauty, etc), but a lot of such criteria is based on actual human perception rules, psychology, and effective methods to convey an emotion, none of that is just random, but the result of centuries long human (artistic) studies.

What you are describing is by definition subjective; IOW, it is on based on or influenced by personal feelings, tastes, prejudices, or opinions. See also among others https://languagetool.org/insights/post/subjective-vs-objective. for more about the difference between making subjective vs. objective judgements.

In short, because subjective judgements are inherently not based on facts there is no general agreement about which criteria to use when deciding if art is good or bad, beautiful or ugly, etc., just opinions about which criteria should be used to judge art, if creating something different by breaking an established rule automatically means it cannot convey emotion as well as something that follows those rules, & so on.

[SrPx]

29 minutes ago, R C-R said:

What you are describing is by definition subjective; IOW, it is on based on or influenced by personal feelings, tastes, prejudices, or opinions. See also among others . for more about the difference between making subjective vs. objective judgements.

 

Nope. Since when techniques are subjective? These are techniques that help creating a better perspective (as in, to paint a landscape or a city scene, building a previous -even if loose, as a good artist eye gets trained with practice- base guidelines of a 2, 3 or 5 vanishing points perspective), better built figures (through solid knowledge and know-how of anatomy drawing (and that is as objective as it can get), shading/lighting, etc), better contrast, better composition (it is studied what the brain (generally) finds more compensated in a composition, this is not "sci fi")... even techniques to generate certain emotion in the user, like psychology of color (studied together with theory of color). These all are studied (and tested) techniques (through centuries) to lead to generate in the viewer a certain psychological reaction or sensation. Indeed, in this is what the advertising language is based on.  This is a key part of what an artist is taught in art schools/college and learns through decades with good books and documentation about all these matters. There is a lot of method in that, which has been tested again and again. There are dense studies about human perception which are pretty objective, (the Gestalt Principles for perception, all the techniques for painting, drawing are in a strong way orientated towards that, color and composition in relation with psychology, etc, etc) based on studies and facts.  The subjective part is that, a part of the art process, not all of it, by any means.  But okay if you don't want to understand/admit that.

[R C-R]

39 minutes ago, SrPx said:

Nope. Since when techniques are subjective?

It is not the techniques used to create something that are subjective. It is the judgements about what makes something better, more beautiful, etc. based on them that is. 

43 minutes ago, SrPx said:

This is a key part of what an artist is taught in art schools/college and learns through decades with good books and documentation about all these matters.

So are you of the opinion that if someone does not attend art schools and/or spend decades studying documentation about such things then they cannot create beautiful or powerful, emotionally evocative artwork? If so, I cannot agree.

[SrPx]

Just now, R C-R said:

It is not the techniques used to create something that are subjective. It is the judgements about what makes something better, more beautiful, etc. based on them that is. 

So are you of the opinion that if someone does not attend art schools and/or spend decades studying documentation about such things then they cannot create beautiful or powerful, emotionally evocative artwork? If so, I cannot agree.

No, that is not exactly the point. There are nuances, let me explain (might we reach some middle ground. But if, not, that's fine, too).  First, artistic expression is IMO beneficial for any human being, if having the desire for such expression (and some don't have it because they never tried, though. Others just really don't feel the need). Like it is also singing or dancing. Almost therapeutic, I would say (I have read it is, indeed, in a bunch of cases).  Anyone can express feelings, emotions, provoking ideas, or simply depict something that they (and maybe others) consider beautiful, etc, through any medium, written words, music. The output of that, anyone can find it beautiful or provoking in any certain way. But like with any form of knowledge, artistic expression (music, writing...), going through well tested methods, good techniques, a person can become a lot more effective in conveying such emotions in others. Like (it's just a metaphor, bear with me as it does not apply perfectly), if I try now to play a song with a violin, my intentions to play something harmonious and beautiful with it can be sincere and intense, but the result for other people's ears might not be so much (or even a pain), as I have never learned a thing about violins, or music, in general.  To the point that someone with years of art learning (painting, drawing, theory, etc) and practice can and surely will become a lot more effective in generating that powerful emotion in others. Even more, in feeling it themselves, which, in a way, is the most rewarding. 

And nope, it is not needed to go to college, as most documentation was already available (and it is yet) in great books in public libraries since many decades (when I actually started learning), or can be purchased for very little, but now internet makes it a hundred times better, easier, and even more, relatively cheap: very affordable courses can put some one in the same path with enough years of practice (practice and constant effort is key). Still, going to Fine Arts college is a great advantage, but I can firmly say that considering the number of my colleagues in the uni that were only interested in "getting a title", that anyone with strong interest, patience, resilience and mostly with good books and video courses (yep, better both things), could very well end up with a much higher skill than most of my colleagues for less money than today most hobbies require, video games included.

But the advantage (just like in science!) of centuries of studies collected in books (even when the cheapest, less recommended art book writers are just repeating techniques which they surely don't know where they came from, or only know barely the practical recipe)  and taught in affordable video tutorials (but good ones! There's a lot of very bad material out there that can do more harm than good) puts the artist going though such (long, I admit it) path in a much better position to build something that aligns with this person's dreams, objectives for creation, and also is more likely to produce the desired emotions in the viewers (this is key). Just because indeed there are techniques for this, just like for every other parcel of knowledge. So, does a prepared artist have the same chances of conveying such emotions in others, or of being as effective in building certain imaginary worlds than a person who picked up a brush for a first time and hasn't read anything about the matter? Well, the first one has more chances (quite more). Is it impossible that a total newcomer to art would produce something beautiful or replicate what is in this person's imagination in an effective way? Nope, is not impossible (there are just quite fewer chances). And in a competitive world, I would say that it is quite more likely that the art of the years long prepared one can match certain criteria (which do exist).

As for the first point, the judgement (ie, an art jury in a contest) is indeed based on the techniques, as it is known that with those techniques is how that effectiveness and "beauty" is better achieved. It all forms a puzzle that a good artist (or an art critic) can establish if it finally works in achieving the goal of conveying such emotion or transmitting a certain idea or sensation effectively enough. Again, because both the criteria and techniques are based on solid, tested aspects of psychology, which is a science, but also on perception (well, it's actually part of it) and art theory, among several other things.