The primary objective for this update was to architect a paradigm shift from a linear generative pipeline to a nonlinear, interactive sound design environment
System Architecture & Implementation of Interactive Components
The existing pipeline, comprising image analysis (object detection, semantic tagging), importance-weighted sound search, audio processing (equalization, normalization, panoramic distribution based on visual coordinates), and temporal randomization was extended with a state-preserving session layer and an interactive control interface, implemented within the collab notebook ecosystem.
Data Structure & State Management A critical prerequisite for interactivity was the preservation of all intermediate audio objects and their associated metadata. The system was refactored to maintain a global, mutable data structure, a list of processed_track objects. Each object encapsulates:
The raw audio waveform (as a NumPy array).
Semantic source tag (e.g., “car,” “rain”).
Track type (ambience base or foreground object).
Temporal onset and duration within the mix.
Panning coefficient (derived from image x-coordinate).
Initial target loudness (LUFS, derived from object importance scaling).
Dynamic Mixing Console Interface A GUI panel was generated post-sonification, featuring the following interactive widgets for each processed_track:
Per-Track Gain Sliders: Linear potentiometers (range 0.0 to 2.0) controlling amplitude multiplication. Adjustment triggers an immediate recalculation of the output sum via a create_current_mix() function, which performs a weighted summation of all tracks based on the current slider states.
Play/Stop Controls: Buttons invoking a non-blocking, threaded audio playback engine (using IPython.display.Audio and threading), allowing for real-time auditioning without interface latency.
On-Demand Sound Replacement Engine The most significant functional addition is the per-track “Search & Replace” capability. Each track’s GUI includes a dedicated search button (🔍). Its event handler executes the following algorithm:
Tag Identification: Retrieves the original semantic tag from the target processed_track.
Targeted Audio Retrieval: Calls a modified search_new_sound_for_tag(tag, exclude_id_list) function. This function re-executes the original search logic, including query formulation, Freesound API calls, descriptor validation (e.g., excluding excessively long or short files), and fallback strategies—while maintaining a session-specific exclusion list to avoid re-selecting previously used sounds.
Consistent Processing: The newly retrieved audio file undergoes an identical processing chain as in the initial pipeline: target loudness normalization (to the original track’s LUFS target), application of the same panning coefficient, and insertion at the identical temporal position.
State Update & Mix Regeneration: The new audio data replaces the old waveform in the processed_track object. The create_current_mix() function is invoked, seamlessly integrating the new sonic element while preserving all other user adjustments (e.g., volume levels of other tracks).
Integrated Feedback & Evaluation Module To formalize user evaluation and gather data for continuous system improvement, a structured feedback panel was integrated adjacent to the mixing controls. This panel captures:
A subjective 5-point Likert scale rating.
Unstructured textual feedback.
Automated attachment of complete session metadata (input image description, derived tags, importance values, processing parameters, and the final processed_track list). This design explicitly closes the feedback loop, treating each user interaction as a potential training or validation datum for future algorithmic refinements.
I recently had the opportunity to visit the Tolkien exhibition in Trieste. The exhibition was dedicated to the life and work of J. R. R. Tolkien and showcased not only his literary world, but also the huge cultural impact his stories have had to this day. I was particularly impressed by three large walls that were completely covered with bookshelves. They contained numerous editions of The Lord of the Rings and The Hobbit from all over the world in a wide variety of formats and designs. (It would be interesting to look at the book covers from different countries and analyze the background and cultural influences on the illustrations.)
At the end of the exhibition, there were also many illustrations by artists and fans who were inspired by Tolkien’s stories. These works ranged from classic drawings to modern interpretations and impressively showed how differently the same world can be visualized.
And again Books
All of this reminded me of a bookshelf at my parents house. For years, it has held several illustrated books dealing with conceptual art and illustrations for the Lord of the Rings and Hobbit films (of course Harry Potter as well). Now, during the Christmas holidays, while I am back at my parents’ house, I went to this shelf and took another look at the books.
One book in particular caught my attention: “The Hobbit: The Desolation of Smaug Chronicles – Concept Art and Design,” published by Hobbit Press. The first edition was published in London in 2013. The main author is Daniel Falconer, supplemented by forewords and texts by other authors and contributors who were directly involved in the film production.
Concept Art at Middle-earth
Conceptual art plays a central role, especially in film adaptations of books. As already mentioned in the blog post about the Harry Potter exhibition in London, conceptual art forms the visual bridge between text and film and helps to translate abstract descriptions, moods, and places into specific images. This work is particularly sensitive and challenging when it comes to a literary source such as The Hobbit, which many readers already associate with their own inner images.
In the case of The Hobbit films, the screenwriters provided numerous ideas that were strongly based on the book. At the same time, The Lord of the Rings already provided existing locations, characters, and visual rules that had to be followed. This made it all the more important to develop the new locations that play a major role in The Hobbit for the first time including Mirkwood (Düsterwald), Esgaroth the lake-town (Seestadt), and Erebor, the Lonely Mountain.
Weta Workshop, which also co-published this book, played a central role in this process. Weta Workshop is a New Zealand special effects company founded in 1987 and based in Wellington. It provides services for film productions, including design drafts, makeup effects, props, creatures, puppets, miniatures, models, and large sculptures. The company also produces merchandising items. For The Hobbit trilogy, Weta Workshop was instrumental in the visual development of characters, costumes, environments, and fantastical creatures, working closely with the director, screenwriters, and other creative departments.
Back to the Book
The illustrated book is structured along the sections of the film and shows the concept art for specific scenes, locations, characters, costumes, and mythical creatures. At the time of publication, many of the artists involved had already been working on The Hobbit trilogy for four to five years. Every day brought new challenges, as concept designer Alan Lee describes in the book’s introduction.
“The book shows only a relatively small part of the enormous amount of work that went into the visualization and production of The Hobbit: The Desolation of Smaug,” says Lee.
The creation of Beorn
The first section of the book is about Beorn’s house. Beorn is a man who can transform himself into a bear. According to the author, this character posed a particular challenge, as he had to be portrayed convincingly in both human and animal form, visually credible but at the same time consistent with Tolkien’s description and the existing imagery of Middle-earth.
I am particulary interested in how artists create fantasy characters, especially those with an ambiguous or hard-to-define personality. Having read the book as well, I can imagine that Boern was one of the most difficult figures to design, because its never entirly clear wheater he is good or bad. His bear form in particular needed to appear frightening and monstrous.
The many pictures in the book are showing how difficult is was, and how many attempts they needed to finding a clear direction. I think its always really calming when you see that talanted artist like the one who contributed here in the films are also struggeling sometimes. Its interesting to see how it works on big sets like this. Many artists work on the same subject at once, each starting with a different vision, and then gradually discussing, refining, and combining these ideas.
John Howe mentoined that for two weeks he made a scribble of boern on his diary every evening. Each day he tried a new approach. Some of these drawings can be seen on the right side of the upper image. For me, this highlights how important it is to keep going, to think differently, and to have the courage to experiment and try things out.
After large number of artworks had been created and presented to the Weta Workshop, the team began developing 3D Modells and exploring anatomical possibilities. Their aim was to push the design in a bit more brutal and extreme direction, which can be seen on the left page of the lower image. In the end, all these ideas come together, resulting in the creation of a truly unforgettable character.
The Elevnking’s Hall
Besides character design, I can imagine designing settings almost as difficult, maybe even more so. Because, as mentioned in the book, you have to think about so many things like light setting, perspective and the story behind it. Across eleven pages, the book shows how the Elvenkings Hall and the Throne room were created. These are two of my favorite locations in the movie because they combine a sense of magic with a darker atmosphere. According to the Conceptartists, one one the Challanges was to continue the Mirkwood but in a more magic way. They wanted to build an entire cave system out of trees that looks very natural.
In the images above you can see a few of the artworks they did. Another important thing in the creation was the deepth the rooms should have. Gus Hunter, one of the designers at Weta Workshop, worked with many columns, trees, and roots to shape the space. When thinking about the lighting of the hall, he came up with the idea of using stalactites in which the elves could place oil lamps. In addition, he used water reflections to help illuminate the entire setting.
To come to an End
Overall, it is incredibly interesting to see how these different departments and artists work together to bring such complex worlds to life. What I have written about here is only a very small glimpse into the book, which itself is just one part of a much larger series documenting the creative process behind The Hobbit. Once again, it highlights how essential concept art and strong visual storytelling are for building believable and immersive worlds. This topic is especially important for my future academic work, as it directly connects to my interest in the visualization of storytelling and in creative development processes more broadly both of which I eventually plan to explore further in my master’s thesis.
Visiting the photography exhibition of Milorad (Milan) Pešić in Cetinje felt less like attending a cultural event and more like stepping into a collective family album. As a girl born and raised in Montenegro, I did not walk into that space as a neutral observer. I walked in carrying memories, inherited stories, and a deep emotional connection to the landscapes and faces that Pešić captured through his camera.
Cetinje itself already carries a special weight for me. Old royal capital has a quiet dignity, a sense of endurance that mirrors the Montenegrin spirit. Seeing Pešić’s photographs displayed there felt right — as if the city and the images were speaking the same language. From the very first photograph, I felt recognition. Not just recognition of places, but of emotions. These were not distant historical documents. These were scenes I felt I had lived, even when they belonged to a time before I was born.
What moved me most was how familiar everything felt. The faces in Pešić’s portraits reminded me of my grandparents, my neighbors. The men — strong, reserved, shaped by the land — carried expressions I have seen my entire life. There is something uniquely Montenegrin in the way people stand, look, and exist in space, and Pešić captured that without exaggeration or romanticism.
As a woman, I found myself paying special attention to the images of women in the photographs. They are rarely posed or idealized. Instead, they appear real — working, waiting, carrying burdens both physical and emotional. Their strength is quiet, almost understated, yet undeniable. Looking at them, I felt a deep respect and a sense of continuity. These women are the reason our traditions survived. They held families together, preserved customs, and endured silence when words were not allowed.
The landscapes, too, stirred something deeply personal. Montenegro is a country of extremes — harsh mountains, calm seas, isolated villages, and narrow streets filled with history. Pešić’s photographs do not simply show these places; they remember them. The mountains feel heavy with time, the villages feel intimate and vulnerable, as if they could disappear if not preserved through images like these. Standing in front of those photographs, I felt gratitude that someone cared enough to document our world before it changed forever.
What makes this exhibition especially powerful is its honesty. Pešić did not try to beautify poverty or dramatize hardship. He simply observed. And in doing so, he allowed dignity to emerge naturally. His work reminded me that our history is not only written in books or political events, but in ordinary lives — in daily routines, in work, in silence, in perseverance.
Leaving the exhibition, I felt both proud and emotional. Proud of where I come from, and emotional because so much of what I saw belongs to a Montenegro that is slowly fading. Yet thanks to photographers like Pešić, it is not lost. It lives on — not as nostalgia, but as testimony.
This exhibition was not just about photography. For me, it was about identity. About seeing ourselves clearly, without filters, and accepting both the beauty and the hardship that shaped us. As a Montenegrin woman, I left with a renewed sense of connection — to my past, my people, and my place in this long, quiet story we call home.
Klanglicht 2026 war für mich ein richtig angenehmer Startpunkt für meine Recherche. Es war zwar erst mein zweites Klanglicht, aber diesmal bin ich ganz anders hineingegangen. Ich habe mir bewusst Zeit genommen und versucht, nicht nur zu konsumieren oder „schön zu finden“, sondern genauer zu beobachten, wie Licht, Sound und Raum eigentlich Emotionen erzeugen. Genau das ist ja auch der Kern meiner Arbeit: Erinnerungen spürbar machen – nicht nur zeigen oder erklären, sondern fühlbar werden lassen.
Was mich bei Klanglicht besonders fasziniert hat, war diese Mischung aus Staunen, Ruhe und gleichzeitig einem totalen Überwältigtsein. Man bewegt sich durch die Stadt, die man eigentlich gut kennt, und plötzlich wirken Straßen, Plätze und Gebäude komplett anders. Vertraute Orte verlieren kurz ihre Alltäglichkeit und bekommen etwas Fremdes, fast Zeitloses. Das hat mich stark daran erinnert, dass Erinnerungen nie fix oder abgeschlossen sind. Sie verändern sich ständig – je nachdem, aus welchem Blickwinkel man schaut, in welcher Stimmung man ist oder welche Atmosphäre gerade entsteht.
Viele Installationen haben genau dieses Gefühl verstärkt. Sie haben nichts „erzählt“ im klassischen Sinn, sondern eher Räume geöffnet, in denen man selbst etwas gespürt hat. Ich hatte oft das Gefühl, dass nicht das einzelne Werk im Vordergrund stand, sondern das Zusammenspiel aus Licht, Klang, Architektur und den Menschen, die sich darin bewegen. Dadurch wurde mir klar, wie sehr Erinnerung auch etwas Situatives ist – etwas, das im Moment entsteht und nicht einfach abrufbar ist wie eine Datei.
Ein Moment, der mir besonders hängen geblieben ist, war eine Projektion, die wie schwebende Lichtpartikel gewirkt hat, begleitet von einem extrem minimalistischen Sound. Es war nichts Lautes, nichts Dramatisches. Viele Leute standen einfach still da, teilweise minutenlang. Niemand hat geredet, niemand hat gefilmt. Das Ganze hatte etwas sehr Intimes, obwohl man von fremden Menschen umgeben war. Es war fast wie ein kollektives „In-sich-Hineinhorchen“. Genau solche Zustände finde ich extrem spannend, wenn es um neue Formen der Erinnerung an Verstorbene geht: Situationen, die automatisch entschleunigen, ohne dabei traurig oder kitschig zu wirken.
Mir wurde bei Klanglicht auch sehr bewusst, wie stark multisensorische Gestaltung Emotionen beeinflussen kann. Licht kann warm oder kühl wirken, Geborgenheit erzeugen oder Distanz schaffen. Klang kann Nähe herstellen oder einen Raum aufspannen, der fast leer wirkt. Und zusammen können Licht und Sound eine Stimmung erzeugen, die ganz ohne Worte auskommt. Das ist für mein Thema besonders relevant, weil klassische Trauer- und Erinnerungskommunikation oft sehr sprach- und bildlastig ist. Man erklärt, beschreibt, zeigt Fotos. Vielleicht müsste man Erinnerungen viel stärker erleben lassen, statt sie zu erklären oder zu archivieren.
Das Festival hat mir außerdem gezeigt, dass Emotionen nicht nur im Inneren einer Person entstehen, sondern stark vom Raum und von anderen Menschen beeinflusst werden. Die Stimmung zwischen den Besucher*innen, das gemeinsame Schweigen, das langsame Gehen – all das trägt zur Erfahrung bei. Erinnerung ist damit nichts rein Privates, sondern etwas, das auch kollektiv entstehen kann, ohne laut oder aufdringlich zu sein.
Was ich vom Festival konkret mitnehme: -Erinnern kann interaktiv sein, ohne dass man aktiv etwas bedienen oder steuern muss – manchmal reicht schon Bewegung, Licht und Atmosphäre. -Emotionen entstehen oft im Raum zwischen Menschen, nicht nur in ihnen selbst. -Erinnerungsdesign könnte viel stärker mit Raum, Stimmung und Zeit arbeiten, statt sich nur auf Objekte, Texte oder Bilder zu konzentrieren.
Klanglicht war für mich damit weniger ein Event und mehr eine Art Denkraum – und ein sehr stimmiger Einstieg in meine weitere Recherche.
Vor Kurzem war unser Studiengang im CoSA – Center of Science Activities in Graz. Ich war ehrlich positiv überrascht – so viel “Interactionnn Design!”
Ich verbinde Museen eher mit stillen Räumen, mit Wandtexten, Abstand halten und „Bitte nicht berühren“ ODER aber mit einem Kinderspielplatz, mit Bällen, Games und angerotzten Spiel-Stationen. Das CoSa war irgendwo schön dazwischen und man konnte selbst die unmotiviertesten Kolleg:innen irgendwo Dinge ausprobieren sehen.
Sehr „Interaction Design“
Was sofort aufgefallen ist: (Fast) alles ist interaktiv. Ich war beeindruckt und überrascht.
Dadurch fühlt sich der Museumsbesuch weniger nach „Fakten lernen“ an, sondern mehr nach Entdecken. Man bleibt automatisch neugieriger, aufmerksamer, aktiver – man will wissen die die Installation funktioniert, „was man da machen kann“ und lernt ganz nebenbei irgendwas über Technik, Bakterien und Co.
Aber auch bei interaktiven Exhibits gilt es die Aufmerksamkeitsspanne und das Interesse der Zielgruppe miteinzukalkulieren… und eine Ausstellung zu Finanzbildung kann eine Bande Student-innen meist nicht mit der Spannung in der Info überzeugen. In der Theorie klang das Konzept eigentlich relativ vielversprechend: Quizze, Bildschirme, Punkte, Simulationen.
In der Praxis zeigte sich aber: Interaktiv heißt nicht automatisch verständlich oder spannend.
Teilweise waren die Screens überladen mit Informationen. Manche Stationen dauerten zu lange oder fühlten sich eher wie ein Test als wie ein Erlebnis an. Man klickte sich durch nicht um die Antworten kennen zu lernen sondern um „weiter zu kommen“.
Ein Thema wird nicht automatisch zugänglicher, nur weil man es „gamifiziert“. Interaktion muss inhaltlich, formal und dramaturgisch zusammenpassen.
Andere Bereiche im CoSA haben das deutlich besser gelöst. Besonders spannend fand ich den Technikbereich, in dem man sein eigenes Fahrzeug zusammenstellt und anschließend in einer Simulation testet. Die Ai-Station von MoYa, die Skizzen (von Kindern etc) scannt, erkennt und zb photorealistisch macht
Der Vibe muss passen
Was CoSA besonders gut schafft, ist eine Atmosphäre, in der man wieder ein bisschen Kind sein darf. Neugierig. Spielerisch. Fehlerfreundlich.
Diese Haltung finde ich unglaublich inspirierend – auch für unsere eigene Arbeit. Gerade im Motion Design oder in der Animation geht es oft um Präzision, Perfektion, Timing. Aber Kreativität entsteht selten aus Druck. Sie entsteht im Ausprobieren.
Der Museumsbesuch hat mich daran erinnert, wie wichtig spielerische Räume sind – analog wie digital. Vielleicht brauchen wir mehr „Playgrounds“ für Kreative. Räume, in denen das Ziel nicht Produktivität ist, sondern Neugier.
During Klanglicht, I didn’t only get inspiration from other installations that were spread around Graz, but especially from the one I was working on myself. Being part of the project “Vertigo” changed the way I experienced the festival, because I wasn’t only visiting the installations, but actively working on one of them. Our installation took place in a church, where we built a huge tower made out of LED panels. Visitors were invited to sit down and watch minute-long animated shows. Each group created a seamless animation combined with sound, guiding the visitors through a topic. Even though all animations were very different, they were connected through the same structure and space.
My role in the project was as a media student, so I was part of the animation team. I searched for different ways to show the storyline and to use sound together with visuals by working with shapes and videos. A big part of the process was trying things out, failing, and then testing everything again directly on the final installation. Timing was very relevant, because the sound and the animation only worked well together when they were perfectly aligned. Even small changes in rhythm could completely change the effect. Within my group, we focused on the topic of urbanisation. We wanted to show, not just with our visuals but also with the sound, that more and more urbanisation is taking away from nature and with that also from us. We worked a lot with colors and rhythm to show the different parts of our animation. The city was shown as cold, hard, and very bright, using colors that were almost neon. This was meant to represent a new, superficial world and was supported by loud construction sounds. Nature, on the other hand, was calmer and a bit warmer. This contrast helped to clearly distinguish between the two parts within our storyline.
We also worked with building and destroying elements within the animation. At one point, a tower was built up and then destroyed again. The impact of this moment was made stronger through sound. Sometimes we also used black screens or short pauses, which helped to create an emotional impact and made the moment feel more intense. I also found it very inspiring to see what the other groups had done. One group showed the topic of love only by using colors and shapes that once moved in sync and then slowly left each other again. This project showed me what power simple shapes can have and how different the effect can be depending on movement, timing, and surface.
The installation would not have worked the same way on a normal screen. The LED panels, their vertical construction, and especially the space of the church had a big influence on how everything was perceived. The space also had a strong effect on the sound, making the whole installation feel more immersive. This project really inspired me and also influenced how I think about my master’s thesis. I loved that the work was abstract, but still very specific in the way it showed a storyline so clearly. It made me realize that it’s possible to tell a strong story without being literal, just by working with shapes, rhythm, color, sound, and timing.
AI is no longer limited to creating images. Today, it is also possible to animate illustrations. Short videos, subtle motion, and atmospheric loops can be generated from a image or prombt. Since I had never animated illustrations with AI before, I decided to try it out and see how it works in practice. i knew that its possible to create illustrateded animation videosfrom the scratch, but I am intrested in using already existing illustrations and animating them. I heared it works best for minimalistic, subtle animations. That’s why I chose my illustration, which I’ve used once before. I think it has the potential to be pretty good. This blog post is an exploration and an experiment with AI and based on the previous post.
Why I Wanted to Try AI Animation
As an illustrator, animation has always felt like a separate skill set. Traditional animation takes time, planning, and technical knowledge. AI promises a faster and more accessible way to bring illustrations to life, even without animation experience. Sometimes even small animations enhance an illustrations more, especially for soocial media.
Over all I was curious to find out:
how much control I actually have
how natural the movements look
whether AI animation feels like a useful tool or just a visual effect
which Generator is the best
How AI Animation Works (as a user)
Most AI animation tools work with an image-to-video approach. You upload a finished illustration and describe how it should move. The AI then creates a short animated clip based on your image and text prompt. Typical movements include flowing smoke, water, light, or subtle object motion. These animations are usually short and loopable, making them suitable for social media, websites, or mood visuals.
Let’s Start Animating
For my first tests, I decided to focus on tools that are commonly used (and free) for image-to-video animations. I let Chat GPT write the prombt for my imagined animation. Here how it turned out and the picture an want to use, which you may already know from my previous article:
Soft, slow animation. The pink smoke gently flows out of the teapot spout and moves upward in a smooth curve. The fish inside the smoke slowly swim forward. Calm, dreamy motion, no camera movement, subtle loop.
It was not easy to find Generators that are completly free. I just can just Firefly because I have an Adobe Subscription but there exclusive material as well. The good thing on Firefly is, that you can choose between a lof of diffrent models. Most websites require you to purchase credits, but there are some models where you have a few credits available in the free plan as Leonardo Ai or Stable Diffusion. However, I assume that you will achieve the best results with paid models.
Firefly Veo 2: You just can create 9:16 or 16:9 which is very limiting, just the fishes are moving, even though one of the fishi sis turning weird, I like the flowy and slow motions
Stable Diffusion: I tried Stable Diffusion for the first time and I have to admit I am very happy with this result, the movements of the fish are really realistic and I like how the smoke is slightly moving in the background, it looks very magical
Firefly Ray 3 HDR: I like how the smoke is slowly moving upwards and how the fish are floating around it, I dont get why the camera is mocing upwards cause there is no information about it in the prombt but I kinda like it, what’s really strange are the colors and how gray and dark they’ve become.
Firefly Pika 2.2: Really dont know whats going on there
Leonardo AI: I had high hopes on that generator but it turned out a bit weird, i like how the smoke is flowing, but everything is moving way too fast
Firefly Veo 2
Stable Diffusion
Firefly Ray 3 HDR
Firefly Pika 2.2
Leonardo AI
I think the best result so far came from Stable Diffusion. That’s why I wanted to refine the prompt and make it more precise, to see if I could achieve an even better result, even though I already really like how it turned out. Here’s what I got.
Unfortunately, I couldn’t manage to keep the fish inside the frame. Aside from that, I really like the movements inside the teapot, especially the gentle water motion and the way the small fish swim. If it were possible to combine these movements with the version shown above, that would be my ideal result.
Final Thoughts
Animating static illustrations with AI is something I find really interesting. Compared to fully creating illustrations with AI, I like this approach because it lets you add motion to existing artwork and create cool results very quickly. You don’t need any After Effects skills, and traditional animation usually takes a lot of time.
Especially for social media or small animated pieces, AI animation is a great option. You can get something moving without a complex workflow. It could also be interesting to animate individual layers separately and then combine them, which might lead to even better and more controlled results. Prompting is still challenging, especially when you have a very specific idea in mind, but overall it’s an exciting and accessible way to experiment with animation.
Before the break I met with Birgit to talk about my master’s thesis. I had a rough idea what I want to do for my thesis since end of November, but nothing really concrete yet. To be honest I also had some doubts about the topic every other day if this is really what I want to do. I thought a lot about the topic and the project, brainstormed, researched but somehow, I circled around the same project ideas over and over again. While this is probably a normal part of the process, I still felt stuck. So, the meeting with Birgit came at a very good time for me, since I wanted to move forward with my idea and get more concrete.
One thing that felt quite clear to me already was the general direction for my thesis. I want to design something that embraces low-pressure creativity, fun, a bit of uselessness, and something whimsical. I feel like we are missing all of these sometimes during our everyday life as creatives. I see this as a counter to the productivity- and hustle-culture we are currently live in. It feels like everything needs to be useful, perfect and efficient. With my thesis I want to explore the opposite: creating a space where people can simply enjoy the act of creating without goals, pressure and expectations.
However, my intention felt clear, I struggle with what the creative space/playground could be. What should be on the website? What do people do there? How does it look like? After pitching her my idea she gave me a historical context I didn’t think of yet. She told me about a the time where the internet was full of so-called “useless” or one-purpose websites, especially during the time of Flash websites. These sites didn’t try to solve a problem or be efficient; they simply existed for fun, surprise and a little confuse.
One example she showed me was the, back in the days very famous, Hamster Dance Website (http://www.hamsterdance.org/hamsterdance/). The website doesn’t really do any useful it is basically just a loop of animations of a hamster with music (unfortunately the sound somehow doesn’t work, but the version Birgit showed me had sound). There is no goal, it doesn’t lead to anything and there is no productivity value and yet it is very joyful and funny. This reminded me that the internet hasn’t always been about optimization, metrics and productivity. It was playful, strange and delightfully pointless. So, I am going to take a deeper look at the history of websites and what was already out there.
Another aspect Birgit told me I should consider is the time factor. She suggested that the webspace or the content of it should not be available all the time. Instead of it being constantly available, it should be available only temporarily, for example 24 hours, before disappearing or changing into something else. That is the same concept as BeReal follows, people can take a snapshot of what they are doing now, once every 24 hours. This limitation creates presence and urgency, but without the pressure to be perfect. This temporal aspect could reinforce the idea of low-pressure creativity: you show up, you play, you create and then it’s gone. You create just in the moment, there is no way to iterate, optimize or monetize.
Even though I still don’t have an exact idea of what the webspace should be like, the talk with Birgit gave me new insights and impulses for the next steps.
AI was used to check spelling and grammar and better clarity.
Artificial Intelligence is currently one of the most discussed and controversial topics in the creative industry. Especially for illustrators, AI raises many questions, fears, and also curiosity. Some see it as a threat, others as a useful tool. Instead of taking a clear position right away, I decided to test AI illustration tools myself and see what they can (and cannot) do. The goal of this article is to explore it through a practical experiment and see what models are the best for illustrations.
Between Fear and Opportunity
I have already written a blogpost on this topic. However, for the purpose of completeness, I would like to reiterate the most important points here. AI in illustration comes with many concerns, and these concerns are valid.
On the negative side, many AI tools are trained on artworks without the consent of the original artists, which raises ethical concerns. Many illustrators fear job loss or that their work will lose value. AI also tends to copy existing styles without truly understanding their meaning.
On the positive side, AI can be useful when used carefully. It allows fast idea generation, quick experiments, and can support inspiration and composition during the creative process.
This mix of risks and possibilities is why AI remains such a challenging topic in illustration.
The Experiment: Recreating My Illustration with AI
For this experiment, I used one of my own illustrations, which I created manually in Procreate. This illustration served as the reference and starting point. The idea was to see:
how close the AI results come to my original illustration
how well the style, colors, and mood are translated
where AI performs well and where it clearly fails
From my experiences using AI Genereators, its easier to let AI write the Prombt for the gererated picture instead of write it by myself. So I gave Chat GPT my Illustration and asked him to describe this illustration in form, color an style for an AI Generator Prombt.
This is how it turned out:
A surreal hand-drawn illustration of a ceramic teapot filled with water like a small aquarium. Inside the teapot, tiny fish and aquatic plants are floating gently. From the spout of the teapot, a soft pink stream flows into the air, and two larger pink fish are swimming within this flowing stream. Dark green textured background, soft pastel color palette with turquoise, pink, and muted grey tones. Whimsical, poetic atmosphere, clean linework, subtle grain texture, storybook illustration style, calm and dreamy mood.
Let’s start the AI Competition
For this experiment, I tested several popular AI image generator models mainly from Firefly. Here are the results and my opinion on it:
Firefly Image 4 Ultra: it dosen’t make sense at all, the style is quite mixed up (3d and 2d elements), don’t understand the stain in the backgorund, at least the teapot looks quite nice with the aquarium in it
Chat GPT Iamge 1.5: I think thats one of the best results when it comes to similarity, the style is illustrative and handmade, it actually make sense, but I don’t like the layout it seems kind of lost and random, the fishis are nice
Gemini 3 with Nano Banana Pro: This modell is usually one of my favorites, I like the artsy style, I think its the best result in gerneral which you could actually use for something (or work further on that), It’s all kind of random and not exactly like the prombt but the style fits pretty well
Flux 1.1 Ultra Raw: The only thing I like here is the bold and graphic style, without the fish in the air and the tiny smoke-thingy it could look nice
Runway Gen-4 Image: What the hell is happing here?! But to be honest if there were no weird eel-ghost there, I would choose that as the winner picture. I really like the style, it looks very editoral-ish and artsy, even though its the most diffrent compared to the original one
Leonardo AI: I tried this website for the first time as I read it should work well for illustrations, turns out: not really. I think its together with the Firefly Ultra its my least favorite, the style is kind of nice (maybe for posters) but it dosen’t make sense
Firefly Image 4 Ultra
Chat GPT Image 1.5
Gemini 3 with Nano Banana Pro
Flux 1.1 Ultra Raw
Runway Gen-4 Image
Leonardo AI
I thought this experimetn was really fun, but I am not completley happy with it at this point. So I decided to take my favorite model and work on the gererated picture until I am happy with the result. Since I really liked the Style of the Runway Gen-4 Image Generator I chose to go with that one, even though the compostions is hilarious. But I see the most potentiol here to get a nice and artsy result.
But after half an hour promting by myself and chat gpt, the picture below was the best result I could get. So I gave Runway up again and switch to my usual favorite Nano Banana again.
Runway
Nano Banana
I tried hard to create a nice result, but the one from Nano Banana was the best I could get. Its kinda nice, but the artsy and handdrawn style is totally missing. It always seems easier than it is to prombt something nice, but actually its pretty hard and time consuming. Probably I could get a better result than this, but I dont wanted to waste so much time (and of course energy) to create those images. And I think thats the whole point: It’s kind of fun to prombt something until it doesn’t work out the way we want or it dosen’t match our expectations. It was interesting and fun to create these images and see how differently each generator works, but it was not as enjoyable as painting the illustration myself. In the end all the created images are looking like AI images. I can’t say exactly why – maybe the too clean lines, the over perfectionsim or the plain textures – but you can see it in all of them. They have no soul and no personality. I don’t want to be an AI-hater, because I am sure that we all have to use it in the future, but with real art at least for me it can’t compete and won’t ever.
I am sure that for some tasks or areas, it makes sense, especially if it doesn’t matter much which illustrations are used. It can save a lot of time and produce beautiful results. Perhaps there are also ways to combine analog illustration and AI, which would certainly be interesting to test. But all in all, I believe and hope that human illustration and art will always remain an important and irreplaceable part of our lives.
Because Art without soul is like a book without words.
Following on from my thoughts in my previous blog post on “Sketchbook,” I looked into the topic of AI and illustration in more detail and came across Eileen Isagon Skyers’ TED Talk. I found her ideas very interesting and similar to my own, which is why I would like to share them here.
What does originality mean when machines can generate images?
That’s the question she is asking the audience. Eileen Isagon Skyers explores how AI is changing art and creativity. Her key points are:
Originality may now be collaborative, layered, and concept-driven rather than purely hand-drawn.
AI produces images that are both familiar and uncanny, blurring lines between human and machine creativity.
Artists working with AI often act as curators, shaping output by selecting, refining, and guiding the machine.
Questions arise around authorship, emotion, representation, and bias: who defines originality, and whose stories are told?
Why this matters for Illustrators
Originality in art is changing. It is not just about drawing perfectly but about showing your own voice, making choices, and putting intention into your work. Illustrators today often guide ideas and improve the results instead of creating every detail by hand. What really matters is emotion and story. AI can make pictures, but it cannot capture real experiences or true feelings. It is also very important to think about representation, to consider whose stories are shown, which visual traditions are included, and which are left out.
AI aka creative thief
When talking about the originality of AI artworks, one must of course also ask where AI gets its references from. That’s right – from all of us. Caught up in the AI and art bubble, I came across two more TED Talks that deal with precisely this topic. One by Ed Newton‑Rex on how AI “steals” creative work and another by Melody Liu on the ethics of AI art. Both show that AI can create impressive images, but it raises serious questions about originality, authorship, and fairness.
AI learns from existing artworks, often without permission. This means that styles, images, and ideas created by real human artists are used to train machines that can then produce new work. On one hand, this can inspire new creativity and speed up the illustration process. On the other, it risks taking away recognition, income, and the personal voice of artists.
Another issue is emotion and story. AI can combine shapes and colors perfectly, but it cannot capture lived experience, personal feelings, or the soul an artist puts into their work. Without that human element, art can feel empty. Finally, representation is also a concern: AI reflects the biases of its training data, so some voices and cultural traditions are left out, while others dominate.
How to protect your art
Ed Newton‑Rex also talks about how artists can protect their work from being used by AI without permission. One important step is to be aware of where and how you share your art online. Artists can use copyrights and licenses to make clear how their work can be used. Another idea is to add signatures or branding so people can see which work is original. Newton‑Rex says it’s important for artists to have a say in how their art is shared online, so they can protect both their voice and their income.
My take away
Of course, everyone knows by now that AI steals from real artists, but it’s easy to ignore and these talks really made me realize that I need to keep it in mind before posting anything online, especially on Instagram or Pinterest. My opinion on AI hasn’t really changed since my last blog post, it feels like most people talking about AI in a good way as “art” aren’t really illustrators or working artists. Still, I’m sure there are positive ways to use AI creatively without losing the soul of art, and I probably need to explore that side more. All three TED Talks I watched were really interesting and raised a lot of important questions. There are definitely many more talks out there on this topic for anyone who’s interested. I’m not sure if I want to explore this topic more deeply at this point, but maybe it could be part of a research.
