We create with AI and then feel guilty. Why does it happen?

“Sometimes I feel guilty when I use AI. Even if I’m not cheating, even if I’ve made the effort to think, to correct, to put my own input into it, even if the result is objectively better. I just feel guilty, period.” This is one of many posts that have appeared on Reddit in recent months, a thought shared by thousands of users — often students, but not only. AI guilt has become one of the quietest — yet deepest — side effects of adopting generative tools like ChatGPT, Midjourney, Sora, or Runway. It is a widespread feeling: the perception of “cheating” when entrusting part of your work, study, or creative process to a machine.

This is supported by data: a 2024 Forbes survey by Brigitte Paulise found that 36% of Gen Z feels some degree of guilt when using AI for professional activities. Similarly, a study by Cecilia K.Y. Chan and Wenjie Hu at the University of Hong Kong identified three main causes of this feeling: the fear of seeming lazy or inauthentic, the fear of others’ judgment, and the sense of losing one’s creative identity.

The paradox is obvious. AI offers us extraordinary tools for writing, synthesizing, creating, and designing — yet many of us feel compelled to justify its use. As if we were cheating. As if the machine erased the human part of our work. But is that really the case? One of the deepest roots of this guilt is cultural. From an early age, we are taught that “doing it yourself” has intrinsic value: the sweat on your brow, the effort, the slowness as proof of authenticity.

In his 2008 book The Craftsman, Richard Sennett observes how Western culture has historically constructed an idea of value based on manual labor and time invested, more than on the final result. Within this framework, using artificial intelligence — which accelerates, simplifies, and suggests — can make us feel as though we’re skipping a necessary step. Even when the result is excellent, there’s still that inner voice whispering: “this isn’t all your doing.” The psychological mechanism is very close to what Pauline Clance and Suzanne Imes described in 1978 as impostor syndrome: the belief that one’s successes are not fully legitimate if they result from outside help. This logic is becoming increasingly unsustainable: no one today feels guilty for writing an essay with the help of a spellchecker or for editing a video with digital software. And yet, with AI, the hesitation explodes.

In creative professions, this guilt seems to hit especially hard. Even when an artist openly uses AI, declaring it without embarrassment, the risk is that the public — and online communities — react with hostility, proclaiming “this is not art.” This was the case for singer-songwriter Giuliana Florio, known as GROSE, who released a song created with artificial intelligence that quickly went viral, only to be swept into a social media firestorm. Similar episodes have occurred with projects like The Velvet Sundown, a non-existent band producing fully AI-generated music, which has surpassed one million streams on Spotify. The problem may not be the tool itself, but rather the fact that these productions can compete — and even win — against those created in traditional ways.

This naturally raises a question: if creating — be it music, literature, or visual art — becomes easier, what happens to the industry and to those still striving for success without “shortcuts”? It’s a legitimate concern: lowering technical barriers and production costs risks creating a glut of content, making it harder to stand out. But history shows this is not the first time: the invention of the printing press, the multitrack recorder, and digital editing software all had similar effects. As early as 1936, Walter Benjamin, in his famous essay The Work of Art in the Age of Mechanical Reproduction, explained how each technological innovation destabilizes traditional criteria of authenticity, without erasing aesthetic value. In every era, attention has ultimately rewarded not those who produce the most, but those who strike authentic emotional chords and create lasting connections.

Perhaps the real issue lies in our perspective. We continue to see creativity as an exclusively human trait because it is tied to emotions — long regarded as uniquely human territory. But, as philosophers like Rosi Braidotti suggest in rethinking the concept of intelligence through a posthuman lens, recognizing forms of complexity in machines and in nature, we could do the same with creativity. A post-anthropocentric approach would allow us to see the creative act as a continuum in which humans, artificial systems, and natural processes intertwine.

However, there is an obvious risk: the filter of capitalist hyperproductivity. In a system that measures the value of creativity in terms of output, speed, and profit generation, AI can be seen not as a tool for exploration, but as a means of flooding the market with “optimized” content designed to capture attention and monetize it. In Psychopolitics, Byung-Chul Han denounces this acceleration of production as one of the causes of cultural impoverishment, where quantity replaces meaning and art is reduced to informational merchandise. Yet art, in its essence, does not obey efficiency metrics: its strength lies not in quantity, but in the ability to open spaces of meaning, to make us pause, to challenge us.

In the end, AI guilt is the child of a culture that has taught us to measure the value of work in terms of effort, and the worth of creativity in anthropocentric terms. But we shouldn’t apologize for wanting to work better. We shouldn’t feel guilty for asking a machine for support, if that support has helped us create something that represents us. As Donna Haraway writes in her Cyborg Manifesto, the challenge is not to preserve some supposed human purity of making, but to recognize the hybrid power of creative processes born from the encounter between human, machine, and environment.