What does the use of Generative AI do to our human creative expression? Part 2: the thinking
- Cecilia Scolaro
- 5 hours ago
- 7 min read
Welcome back, I hope that before you read this piece, you have taken part in Part 1 of this Article, the experiment. There I asked you to try a few tasks and see how it made you feel and what it made you think.
I came up with those exercises inspired by some general observations and some pieces of academic research that all point in the same direction:
generative AI mediated creative practices are eroding fundamental skills of human expression, as much as capitalist and colonial systems have been eroding skills for collective creative expression.
My intention with this piece is less about determining whether someone should use AI or not, or even how to use it “ethically”, but rather offering vocabulary and critical tools to (creative) people to examine what happens to their own expression when using Generative AI.
In other words, when you find yourself in the position of choosing to use AI or already in the midst of it, can you tell how your expression is influenced by it? Can you tell what you are losing of your own identity and culture?
My goal is helping you answer these questions for yourself, to yourself and then choosing yourself what to do with it. If you follow my work, you’ll know that I like to challenge the common belief about the neutrality of the tech tools and offer critical and practical tools to people. This won’t be any different, but let’s take a couple of steps back first.
Human expression
Human expression has always been the manifestation of a certain community, a culture, a ‘collective consciousness’. Throughout times, geographies and cultures people have created together or alone, to express and communicate their times, their values, their experiences, their pains, their joys. Many collective practices still thrive around the world and carry on traditions of beauty, spirituality and resistance.
Collective creations are so much more than a moment of shared creativity, they bring together shared values and belief, common memories, collective experience and interpretations. They are the result of collective negotiation among diverse ideas and perspectives. They carry profound connections with the land, the ancestors and the community.
The erosion of collective practices
The colonial and capitalist system we live in spent the last couple hundred years selling us another narrative regarding human expression, the idea of the lonely creative genius, the isolated creator, alone capable of interpreting the world. We replaced collective meaning with individual interpretation, the relationship with our ancestors with the analysis of our psyche, devising “the creative process” as an hero journey of personal struggle, where only a few were able to venture.
This narrative, although certainly prolific, has eroded our capacity of expressing creatively together, negotiating the creative process, and therefore creating output representative of who we are as a people.
The relationship became exclusively between the artist and the object of creation while the community, the ancestors, the land were somehow fading in the background or exclusively mediated by the eyes of the artist. Collective creation persists, but its practice is often relegated to ‘craft’ and ‘folklore’ while individual creation is elevated as ‘art’ and ‘genius’.
What is happening with Generative AI?
Generative AI has entered all realms of expression. Willingly and often unwillingly people are using it to write, draw, design, play music. There is mounting evidence that the use of generative AI is influencing not only how we create, but what we create, who we create as, and even why we create.
It is, at all effects, further eroding our human capabilities of expression by replacing our personal perceptions, interpretations, memories and values with statistically relevant output. After disconnecting us from land, communities and histories, generative AI creates the conditions for potentially disconnecting from ourselves.
Let me share a few pieces of academic research about AI and some common general observations that support this thesis.
AI-Influenced writing
Several researchers are investigating the surge of so-called “AI-Influenced Vocabulary”: the use of words like ‘delve’, ‘underscore’ has spiked across scientific writing within a relatively short amount of time.
Even without scientific research to demonstrate it, any written text you come across these days is showing signs of usage of LLMs: the proliferation of long dashes, the expressions ‘not only…but’, the several ‘in conclusion’ and overall the prosaic quality even mundane simple emails have acquired. A lot of people are opening their Generative AI tool before even figuring out what they want to write about and ‘brainstorm’ with the tool on topics, language and style.
Considering most people are using the same large models, and considering the bias that has been demonstrated these models have towards certain expressions, what effect does Generative AI have on our collective and individual meaning making? How does using AI contribute to an homogenous, tech-driven meaning and eroding the individual intention and the cultural significance behind every word?
Cognitive and memory impairment
A team of researchers at MIT and other Massachusetts institutes have run an experiment on the cognitive and memory impact associated with the use of generative AI. They gave the same creative task to three groups of people: one group was asked to use a generative AI tool, one was given a search engine tool and the last group was asked to work without tools ‘brain-only’.
The results are astonishing: through electroencephalography, interviews and NLP analysis the researchers were able to conclude that the ‘Brain-only’ group had significantly more neural connectivity patterns, reflecting divergent cognitive strategies - while “brain connectivity systematically scaled down with the amount of external support”. In addition, and to me even more alarmingly, among the people making use of LLM only 16,7% were able to quote from the essays they wrote just minutes prior.
How are our own cognitive capabilities impacted when using generative AI? Are we going to remember what we created? Are we remembering what we are learning? What is the point of creating something that is not remembered neither from its creator?
Algorithmic Self
Psychology and philosophy scholars are investigating the impact of interacting with technology who continuously profiles us and projects a certain image back to us. They call it the “Algorithmic Self”. The “Algorithmic Self” refers to a form of digitally mediated identity in which personal awareness, preferences, and even emotional patterns are shaped through continuous feedback from AI systems [...]. The more time we spend in a world where algorithms dictate so much of what and how we know, the more necessary it is that we explore how AI reconfigures not just what we are doing, but who we are doing it as.
How is our own perception of ourselves and therefore how we want to express ourselves impacted by the use of generative AI? How is the output of the AI tools we use shaping our own sense of self? How is it influencing what we are going to create individually and collectively? Last but not least we should ask ourselves: what are the short and long term implications of so many people abdicating their own creative process, their skills as much as we abdicated on so many cultural practices?
There is a phenomenon called skill decay: it describes when we lose a skill because we don’t exercise it anymore. If you grew up before the domain of smartphones you may remember knowing phone numbers by heart, navigating in places with paper maps, without GPS! Those skills were replaced by technology and we lost them. Someone may argue that they were not very significant skills, but what happens when we lose our own ability to express ourselves, to tell who we are, to be creative? Especially at the scale and speed generated by the escalation of AI systems in our lives.
The historical moment we are living in should not be underestimated, we can choose what to keep alive. How are we keeping both collective and individual human expression alive despite the pressure, the ‘inevitability of AI’, the FOMO? How do we prevent the homogenization, the loss of meaning, the abdication of our cultural values caused by the use of generative AI?
We may look for guidance into how so many beautiful practices around the world have been kept alive despite the violent systems that have been actively attempted to erase them. Their survival and evolution was made possible by people, who made deliberate and sometimes difficult choices, decided to learn and pass on to their children, prioritized their collective memories and meanings, and did not cave in the powerful erasing forces of colonialism and capitalism.
They did not do it only out of idealism but because collective creation regenerated them as well, as much as remembering and expressing your place at the beginning of this piece may have regenerated you. Because human expression is something that we need as much as we need to connect with others, water and food. If you are longing for more of that connected feeling, be in touch. We can, together, choose our collective expression.
Note: in writing this article I refrained from using generative AI. It may not be perfect, but it completely represents me, my point of view and my intentions. Most importantly, it does not rely on intellectual property theft and over-consumption of resources.
References
Matsui K. Delving Into PubMed Records: How AI-Influenced Vocabulary has Transformed Medical Writing since ChatGPT. Perspect Med Educ. 2025 Dec 2;14(1):882-890. doi: 10.5334/pme.1929. PMID: 41356414; PMCID: PMC12679996.
Joseph J (2025) The algorithmic self: how AI is reshaping human identity, introspection, and agency. Front. Psychol. 16:1645795. doi: 10.3389/fpsyg.2025.1645795
Kosmyna, Nataliya et al. “Your Brain on ChatGPT: Accumulation of Cognitive Debt when Using an AI Assistant for Essay Writing Task.” ArXiv abs/2506.08872 (2025): n. pag.
Macnamara BN, Berber I, Çavuşoğlu MC, Krupinski EA, Nallapareddy N, Nelson NE, Smith PJ, Wilson-Delfosse AL, Ray S. Does using artificial intelligence assistance accelerate skill decay and hinder skill development without performers' awareness? Cogn Res Princ Implic. 2024 Jul 12;9(1):46. doi: 10.1186/s41235-024-00572-8. PMID: 38992285; PMCID: PMC11239631.
Comments