By 2028, creators working in the audiovisual industry are expected to lose around 21% of their income. The effectiveness of copyright protection in its current form is increasingly being challenged.

https://www.sirp.ee/kolm-pilku-filmindusele-tehisaruga-ja-ilma/

Over the course of decades, composer Arvo Pärt has sent his friends, colleagues, and collaborators distinctive greeting cards to mark special occasions. Each card is built around a sheet of manuscript paper, where the “title of the work” is the name of the person being congratulated.

From this idea emerges a kind of musical code or formula: the greeting itself consists largely of written-out music and text. Pärt often quotes the well-known “Happy Birthday to You” and other characterful musical pieces. And indeed, character is the keyword here. As a young man, while helping his mother in a kindergarten, Pärt would play children's music of varying character on the piano and compose playful children’s songs.

Between 1962 and 1979, Pärt collaborated extensively with filmmakers, composing original music for more than thirty films. Among them are energetic animated works such as “Väike motoroller” (“Little Scooter”), “Aatomik”, “Pallid” (“Balls”), “Värvipliiatsid” (“Colored Pencils”), and “Hiirejaht” (“Mouse Hunt”). Synchronizing music with image was, at the time, a technically complex process. All the more admiration is due for the way Pärt stretched his imaginative wings and created such vivid, illustrative music. In a radio program from that era, he noted that the essence of each film’s music is also connected to the director’s inner world and character. This reveals the level of empathy in Pärt’s collaborations, and it is no surprise that so many film professionals wanted to work with his music. Humor, playfulness, and vivid imagery are therefore also strongly present in this book.

At the same time, these musical fragments also contain qualities so characteristic of the later Pärt: seriousness, reflection, longing, sorrow, and consolation. Reading these greetings and congratulations, one feels almost slightly guilty, as if secretly leafing through someone’s diary. These are intimate gestures of celebration. Gratitude and humility are so clearly present that they can make the reader slightly self-conscious.

The code or formula described here is a good example of how a composer defines his own boundaries. Strict structural thinking, a mathematical framework, a degree of serialism, and early forms of graphic notation could serve as excellent study material for an aspiring composer. From the recipient’s name, those letters are selected that correspond to musical note names. The remaining letters become rests or are simply ignored. In this way, the composer already has material with which to “play,” even if it consists of only a single pitch. Traditional compositional techniques such as inversion and retrograde are employed to create a unique musical “signature” for a name. These greetings can also be seen as sketches for larger works. Every person is, in a sense, a great work of nature.

As mentioned, the book has a cinematic logic: contrasting materials, collage-like structures, abrupt mood shifts, and moments of surprise. The legendary music educator Helju Tauk has noted that she was particularly fascinated by Pärt’s early works because each piece, despite differing compositional techniques, had a clear dramaturgical logic.

Does a musical work exist if it exists only on paper? That is the central question. Fortunately, most of these greetings are also available as audio files on the APK website. One example is greeting no. 24, the “Estonian Wedding Dance” written for the wedding of David and Mirjam James, which could be seen as a new piece for weddings alongside “Ukuaru Waltz”. Here, an Estonian folk *labajalg* waltz meets Mendelssohn’s wedding march in a witty fusion.

It is also worth noting the range of recipients of these greetings over the years. The composer’s cards have been sent to presidents, notaries, major benefactors, lawyers, entrepreneurs, music and art scholars, church figures, musicians, composers, publishers, conductors, photographers, artistic directors, producers, ministers, sound engineers, bankers, archivists, and many more. An impressively diverse company indeed.

Finally, it should be mentioned that the book itself is beautifully designed. Its open binding is covered in fabric, giving it a distinctive and aesthetic presence. It was selected among the “25 Most Beautiful Estonian Books of 2025.” The designer behind this craftsmanship is Angelika Schneider.

In closing, a small appeal to all readers: let us send more greeting cards. They matter.

https://kultuur.postimees.ee/8466476/arvo-pardi-muusikalistest-viguritest-sai-intiimne-raamat

Software can open the door to creativity for more people, but there is a dangerous self-deception hidden here. The formula “no tedious learning, no effort, just click and done” does not amplify creativity; it removes the singular process in which the author steps outside mapped territory, learns and experiences something new, and returns with something unexpected. In the latest installment of Ardo Ran Varres’s interview series, artist Peeter Laurits analyzes the problems emerging from artificial intelligence.

The digital age has profoundly shaken art and photography: technological development suddenly makes it possible to work on entirely new foundations. Is the arrival of AI in the field a natural technical development, or do you see it rather as a fundamental rupture?

There is both chance and inevitability in the development of technologies. Every new invention creates the conditions for a whole series of other inventions and directions of development. The digitization of images and generative software are a natural continuation of the current technological trajectory, but culturally this is a very radical break.

What did the digital turn do?

It changed the form and workflow of visual media. But artificial intelligence is already changing the very logic of creation itself. Generative creation is not simply better paint, a more perfect camera, or the replacement of the darkroom with Photoshop, but a system capable of synthesizing images, imitating styles, and proposing compositions through statistical algorithms. In this process, the artist becomes more of a guide, curator, and editor. The rupture is especially sharp in photography, because photography’s value has so far been connected to some trace of reality. Although images have always been manipulated and staged, there was still something in front of the lens. Now that is no longer the case. An artificial intelligence that has never been outdoors and has never seen anything with its own eyes can generate virtually anything convincingly based on existing images … This raises an entire series of questions about the boundaries of truth and trust. Already now, the line between an artwork and a deepfake is very blurred.

Peeter Laurits: “Legally, the question is whether using someone’s works as training data constitutes fair use or unauthorized exploitation.”

Could you have imagined even ten years ago that the intersection of AI and the creative fields would reach where it is today? Machines were supposed to help people create more easily, not start creating on their own…

All these questions had been in the air for a long time. Technocrats have always believed that art can be automated, that clever machines could produce art themselves. Produce, precisely because of marketing fundamentalism, it seems self-evident that art is a product and entertainment.

I am reminded of Anders Härm’s reaction to the marketing clips for the European Capital of Culture program Tartu 2024, about which he once wrote on Facebook: “Culture is radical, heterogeneous, nihilistic, destructive, ambivalent, intellectual and witty, poetic and political, dark and affective, conceptual and precise, sharp and profound … But not some laundry detergent commercial that can be churned out somewhere in Poland and then shown all over Eastern Europe.”

Art may sometimes have the characteristics of a product, but in essence it is something much more: a question, a proposal, a counterargument, a vision, social glue, an explosion, and a vector toward new developmental paths. The art field is like a conference of new beliefs and visions of development. Artificial intelligences, however, do not create art but produce imitations of its external form. If an image generator were suddenly to awaken from standby mode because of an unexpected idea, begin experimenting with a motif, abandon emotionally failed drafts, seek striking clarity, and strive toward illumination, then I would be willing to discuss whether it creates art. Current generative software merely recombines what already exists and produces statistical transformations of it.

Let us hope AI does not reach awakening and enlightenment. Perhaps arranging zeros and ones into new sequences is actually the universe’s cunning plan to awaken us? For many people, combinatorics itself is already nirvana.

What is deeply unpleasant is that such practices reproduce the notion of art as an entertainment product and pour water onto the mill of project culture. In project culture, the creative process is turned upside down: to receive funding, one must formulate the outcome in an application before the creative work even begins, eliminate risks, and provide answers before the questions. This suffocates revelation, hinders innovation, and traps us in a dead circle of recombining what already exists.

Exactly. It would be rather absurd to imagine, for example, that Konrad Mägi or Eduard Wiiralt would have had to write “creative industry projects.” In music and literature, experts consider it important to distinguish between generative AI and human-made work. Machines have been fed enormous quantities of human-created material without permission or royalties. How are these issues viewed in the visual arts?

It is the same in visual art. AI systems have been given access to the entirety of digitized human experience to date: everything available on the internet. The claim that this is the inevitable development of technology is false, because the way these systems were built was a choice. Different models could have been built instead: licensed datasets, collective compensation systems, opt-in registries, clear provenance tracking … Speed and scale were chosen instead of fairness. The current choice creates legal, economic, and moral problems. Legally, the question is whether using someone’s works as training data is fair use or unauthorized exploitation. Economically, the system extracts value from creators’ work, but the value does not flow back to creators. Morally, the issue is even sharper: what is taken from the author is not merely a single image or text, but their signature style, years of work, and sometimes the market position of an entire profession.

Indeed. Software that can create a “new” musical work from scratch in seconds is described by its developers as giving creative opportunity to everyone, not just the talented or chosen few. Their goal is to expand the music industry to the scale of the video game industry, which is currently at least ten times larger. Passive consumption is supposed to become active “creative” activity so consumers can have “pleasurable musical experiences.” A quick and cheap shortcut to the final result: no tedious music education, no inventing melodies, harmonies, and lyrics, no complicated recording process. Click and done. Do you hear similar technocapitalist arguments in the art world?

Of course. This DIY aesthetic has flourished for a long time. Superficiality, predictability, incompetence, and standardization are inevitable outputs of project culture, and naturally, there are advocates for such aesthetics. It is true that software can open the door to creativity for more people, but there is a dangerous self-deception hidden here. The formula “no tedious learning, no effort, click and done” does not amplify creativity but removes the unique process in which the author leaves mapped territory, learns and experiences something new, and returns with something unexpected. Launching a generator is not creativity but frictionless production, the end result without the journey. It does not transform passive consumption into active creativity but merely creates a new form of consumption. The dream of platform capitalists, expanding the music industry to the dimensions of the gaming industry, says quite a lot. It does not sound like a cultural vision but like a program for market expansion. Music and art become infinitely producible and customizable mood services. From a business perspective, this may sound brilliant, but culturally it is miserable, beggarly.

I understand that at the Estonian Academy of Arts, drawing skills and other classical techniques are no longer essential for admission, unlike in the 1950s when my mother studied graphics there. Does AI “democratize” even the last remnants of craftsmanship?

That is true, and not only at the Estonian Academy of Arts but at many educational institutions around the world — and it is absurd. The manual and bodily ability to reproduce something creates psychological feedback loops, broadens thinking, and fosters creativity more than almost anything else. Drawing is not simply “copying something beautifully”; it trains attention and the ability to see. Proportion, rhythm, light, the body, space, tension, silence, emptiness, fullness — all are part of the perceptual discipline. The more innovative our ideas and visions become, the greater virtuosity is needed to shape them into form.

The question now is what is taught instead of drawing skills. If strong courses in visual thinking, semiotics, compositional perception, art history, media criticism, and ethical and aesthetic understanding replaced them, that could be a positive substantive change. In reality, however, art schools were squeezed into the Procrustean bed of the Bologna Convention’s 3+2 academic model. The volume of the study simply shrank, without anything important being added. In this way, replacing manual skills with rapid generation is not democratization but impoverishment.

What new forms and aesthetics has AI brought — or will it bring — into art?

There are many new forms. First of all, prompt art is work created not manually but through verbal instruction, selection, and iteration. The artist does not make the image but summons, directs, nudges, and filters it like a curator. Another phenomenon is synthetic photography: the result looks like a photograph but is not connected to any real event or object. Here, a new aesthetic tension emerges, in which the image hijacks the authority of photography and pretends to be a trace of truth, without any guarantee of reality. Thirdly, latent collage aesthetics have emerged. Generated images arise from compressing an enormous amount of visual memory, resulting in an overabundant image space that mixes styles and motifs. Not classical collage where the seams are visible, but smooth collage where the seams are hidden. Glitch aesthetics are also popular, emphasizing the oddities of digital imagery: oversized pixels, extra fingers, melting surfaces, dreamlike spaces. This turns digital malfunction into a visual language of its own. But these are mostly external features — explorations of new possibilities.

How do you think the cultural sphere will change in the near future?

I believe that the fundamental ruptures and new qualities are not emerging inside digital art itself, but around it, in other forms of expression and throughout the cultural sphere as a whole. After the invention of photography in 1840, relatively little remarkable happened within photography itself for a long time. The revolutionary change was that, contrary to what technocrats of the time believed — namely, that painting would become unnecessary — painting was liberated from its mimetic function and gave rise to numerous paradigm shifts. Few people at the time could probably have foreseen the coming of Impressionism, Expressionism, Surrealism, or Constructivism. Likewise, I cannot predict exactly how the cultural sphere will change over the next decade.

One tendency already visible in video and sound art is the emergence of processual works: they are not clearly bounded objects but rather flows, systems, or environments — meaningful membranes that change in real time. Here, art resembles a living organism more than a finished artifact. Hybrid forms are emerging in which human-made and generated imagery intertwine so that what matters is how the different layers comment on one another. In this new dialogicality, the artwork is not merely the final result but also the field of tension between human intention and the system's unpredictability.

How are these developments affecting illustrators, designers, photographers, and similar professions today?

As I said earlier, the arrival of artificial intelligence threatens the market position of many professions and radically transforms entire cultural fields. Artists now have to fundamentally rethink their role. That is not necessarily a bad thing. Since the 1980s and 1990s, project culture and mass culture have increasingly produced standardization and self-plagiarism — a depressingly uniform visual field that resembled generative slop long before the first prompt was formulated. The AI enthusiasm that will continue for some time now floods the entire visual space with such a thick layer of statistical paste that it starts triggering a gag reflex. This gives artists an excellent opportunity to foreground what is specifically human and cannot be generated.

So AI helps us bring out our unique and inimitable qualities. But questions surrounding copyright, training data, and authorship remain unresolved?

Our entire framework of copyright and authors’ rights still remains unmapped. This should have been addressed already at the beginning of the last century. Current copyright laws date back to the eras of the French Revolution and the American Civil War. Compared to that time, the entire cultural field has changed beyond recognition, and hopefully the arrival of artificial intelligence will provide a shock strong enough to thoroughly refurnish this outdated legal space.

Recently, courts in the United States have ruled that AI-generated artworks are not eligible for copyright protection.

That is not even a partial solution, because many generated works also contain human contributions. For example, the subject may have been photographed by a human while the background was generated. Moreover, every recent-generation Photoshop file already contains generated elements. On top of that, the rights of authors whose works were used to train AI systems remain unresolved. But this becomes truly complicated. When I think about the data arrays used to “train” me as an author, I realize that my consciousness has absorbed countless images, books, and musical works. Likewise, the authors of all those works relied on countless earlier works, and so on. We all train our consciousness and creativity on the basis of previous cultural achievements. Culture itself is citational and dialogical, and plagiarism is not clearly definable. The idea of the author as the owner of their creation is highly simplified.

I am reminded of an anecdote in which an accomplished cellist draws a low open C note on an expensive instrument, and it sounds magnificent. The room fills with overtones and vibrations. A young composer in the same room, who has managed to write the corresponding note onto sheet music, asks in amazement: “Did I really create that?”

When I ask myself whether I own my consciousness and creativity, my train of thought becomes tangled. It seems the question itself is wrongly formulated. Can I own something that is vastly more complex than I am, that exceeds my dimensions enormously in both space and time, and that obeys my will only indirectly? The same question arises with the concept of land ownership. How can a human own a living environment? Is a dog’s flea the owner of the dog? Does a landowner also own earthquakes? It seems to me that in rethinking the issues of copyright and authorship, we first need to detach authorship from categories of property ownership.

We also cannot avoid the environmental aspect. Every prompt harms nature. What is the current state of affairs in this field, and how dark a future do you think awaits us?

In the article “The Information Catastrophe,” Melvin M. Vopson treats information not merely as an abstract description but as having physical status, allowing very large extrapolations among mass, energy, and bits. The central claim of the work is that humanity currently produces on the order of 10²¹ bits per year, and if digital content continued growing at 20% annually, then in about 350 years the number of bits would exceed the estimated number of atoms on Earth. Vopson argues that in about 250 years, the power required to sustain digital production would exceed the current scale of total energy consumption, and that in about 500 years, digital content would, according to his model, make up more than half of Earth’s mass.

Finally, let us take a practical example. An 11-year-old girl has discovered a great passion and talent for expressing herself as an artist: she sketches, paints, crafts, crochets imaginative toys, designs clothes, and draws distinctive portraits at school fairs. What kind of world awaits her when she becomes an adult and wants to devote her life to this field?

Many people already have a strong gag reflex toward deep-fried statistical sludge. If that 11-year-old girl has managed to escape the addictive disorder of the smart mirror, and she sketches, designs, and crafts, then a bright future awaits her. Conscious resistance to machine-like smoothness and fast art is becoming an increasingly valuable skill. The more flawless, slick, instantly readable images are produced, the more valuable friction, materiality, interruption, and slowness become. Not because of nostalgia, but because of distinction. An aesthetics of origin is emerging, in which part of the work's meaning lies in how it came into being. The process itself increasingly becomes a visible aesthetic and ethical layer. It seems to me that alongside slop and quick solutions, artificial intelligence will also bring protest, fresh winds, and new ontologies into art.

In the first interview of the series, Mati Kaalep answered Ardo Ran Varres’s questions (Sirp, January 9, 2026), and in the second, Rein Raud did so (Sirp, February 27, 2026).

* The Information Catastrophe – published in AIP Advances, August 11, 2020. DOI: 10.1063/5.0019941. At the time of the work, Vopson was affiliated with the School of Mathematics and Physics at the University of Portsmouth.

 

Original article: https://www.sirp.ee/kunsti-regressioon-tehisarupohises-kultuuritoostuses/

Rein Raud: “They always thought that robots would come, start washing the floors, and people would be free to devote themselves to art, poetry, and composing music. On the contrary, it is people who will be cleaning and maintaining things so that the robot can engage in ‘creative work.’”

https://www.sirp.ee/kultuuriline-sulgumine-malu-koloniseerimine-ja-tehisaru-rahvalaulikud-2-0/

Mati Kaalep: “By around 2030–2032, we will likely be looking at the music industry in a fundamentally different way — at how royalties flow, but also at how composers create music.”

https://www.sirp.ee/kas-tehisaru-viib-muusikatoostuse-kollapsini/

A journey of thought into the lives and creative work of Veljo Tormis and Arvo Pärt.

https://www.temuki.ee/2025/11/sona-saab-heliks-loitsuks-palveks-muusikaks-armastuseks/

“Sooner or later, most musicians develop a serious need for music notation software — especially composers, of course. We spoke with composer Tõnis Kaumann and musician and music engraver Jaan Kiiv about the notation programs Finale and Dorico.”

https://www.ajakirimuusika.ee/muusika-ja-tehnoloogia/