Resonance in the smallest of spaces: in the hustle and bustle of Deutsche Telekom's ExplorAItion Days, we sit opposite each other with headphones on. I'm playing the piano, and the longer our eyes meet, the more intensely I feel the connection of this moment.

“I can feel it when I see your fingertips touching the keys,” she tells me afterwards with a smile on her face.

While I wonder whether an AI will replace me, I have Keith Jarrett's *Paris Concert* in my ear – the part where he sighs with relief. These are the moments when he is surprised by his improvisational skills and discovers new worlds of sound that have never been heard before. They are also the moments we talk about because they resonate with us.

The very question of whether humans or AI can create 'better' art testifies to humans’ unique ability to consciously self-reflect.

1:0 for human intelligence

Humans can reflect on their perception – an ability that constitutes their self-awareness and allows them to say “I”. In contrast, AI (still) lacks reflective imagination. It operates based on probabilities, the processing of which does not allow for deliberate control. Humans, on the other hand, can consciously go against probabilities to create something surprising or new. Humans imagine results, which enables them to give their work purposeful coherence and meaning.

This explains why previous attempts to complete Schubert's or Beethoven's 10th Symphony using AI have remained artistically unsatisfactory.¹ Behind the deliberate conception of Beethoven's motifs is a human being with consciousness and intention. Truth cannot be calculated – it must be lived and felt.

But AI has its own strengths – strengths that can complement human creativity rather than replace it.

While humans contribute intention and intuition, AI uses its enormous computing power to open up perspectives that often remain hidden from the human eye.

To create something new, it is usually necessary to understand what already exists. AI's ability to analyze vast amounts of data can reveal patterns and connections humans might overlook. It broadens the horizon for research and inspiration but remains limited by the specifications of its human creators. For example, before data is incorporated into a model, it is filtered – a process that can introduce unconscious biases.

The introduction of prompt engineering has made art creation accessible to more people. Those without the talent to play the piano can instead create a new song through clever prompting. This participatory approach broadens the creative spectrum and democratizes access to art – especially for people who were previously excluded by socio-economic barriers. Nevertheless, AI remains a tool: it does not produce art on its own but serves as a means of initiating creative processes.

Art with feeling

Despite all the technological innovation, the question remains: how does AI influence the artistic process, and can it replicate the physical resonance of a human artist?

Isn't it precisely the sensitive touch on piano keys and the emotional presence of the artist that trigger resonance in the other person?

I imagine an AI robot playing the piano in my place and responding to a listener's tears with its own “reactions”. But this moment would shatter as soon as the one-way street becomes visible: an AI that merely follows its model parameters and whose reaction remains a calculation – without real feeling.

The experience of resonance includes intuition, surprise, experience and vulnerability - aspects that are only part of the human being. Why are Glenn Gould's interpretations of Johann Sebastian Bach's Goldberg Variations considered so unique? It is not only his technical perfection or his innovative and headstrong approach to tradition that gives his art depth, but also his eccentric and unpredictable personality.

The artist's gestures – the visible expression of an inner experience – transfer this passion to the audience. Physicality and the psyche are inextricably linked: Embodiment is not just a method of creating art, but a fundamental characteristic of art itself.

A piece of music composed by AI does not “feel” the resonance of an instrument. The touch that gives expression would remain mechanical. Although AI can imitate surprise, it lacks the underlying meaning and intention. Art is created in an open dialogue between the artist and audience – through unpredictable, deeply human relationships.

The acceleration of creative processes through AI

Working with your body requires effort and time. Works that would have taken months or years in the past are now created in a matter of seconds thanks to AI. But in this acceleration, a crucial dimension is lost: the conscious interpretation of errors.

And it is precisely these “mistakes” that make people in art so unique. Human error is often the key to innovation. An unintentional deviation, such as a “wrong note”, can open up completely new worlds of sound. Although “mistakes” can also occur with AI, it remains incapable of consciously interpreting or creatively reinterpreting them. AI can do many things – but it has no artistic vision of its own.

AI masters the “how” of art with impressive precision, but the “what” – the soul, the deeper experience and the intentional interpretation – remains closed to AI.

What do we learn from the interaction between man and machine?

There is no AI without human intelligence. AI remains dependent on humans, as so far only humans have been able to create original ideas from which AI can draw. Without this continuous enrichment, a feedback loop would arise that would prevent innovation. Without humans, we would have a pure remix culture.

At the same time, new technologies have always triggered artistic developments, for example, the invention of photography in the 19th century led to counter-movements such as Impressionism. What art forms will emerge in response to generative AI art?

Art teaches us how humans bring in emotion, intention and subjectivity – qualities that current language models, for example, do not possess. Many new generative models, on the other hand, increase participation and efficiency and complement human creativity without replacing it. Because emotion is a key difference between human intelligence and AI, the term Homo Sapiens (the rational human being) could usefully be expanded to include Homo Sentiens (the feeling human being). This change of perspective emphasizes the uniqueness of the human being and enables a symbiosis between man and machine that needs to be further explored through art.

Footnote

¹Dirk Kaftan, conductor of the performance of Beethoven's 10th, confirms this insight: “AI processes things that have already happened. It processes the past. The question is whether it creates something original and new that captures a zeitgeist from the human soul and turns it into an unmistakable work of art - that has not been answered with this project. And I have very, very serious doubts about that.” https://www.br-klassik.de/aktuell/news-kritik/kritik-urauffuehrung-beethoven-10-symphonie-kuenstliche-intelligenz-computer-bonn-100.html