It has been a long time since I posted my thoughts on this blog. I was simply overwhelmed by other work. There could have been a solution, though. I could have collaborated with AI to produce drafts of new entries. Should I have done that?
In my role as editor-in-chief of the journal Cultural Science, I strongly endorse experiments in human-AI collaboration in authorship. For example, my colleague Martin Bohle, an associate editor and guest editor of a special issue, published an editorial featuring the title “Martin@ai Bohle.” I have explained the background of this editorial policy in another post on this blog: I regard AI as a manifestation of Marx’s “general intellect,” which enables individual humans to access the full extent of human knowledge, as it is available on the internet and/or is used to train AI. This means that, together with the human role in prompting and guiding the AI in producing texts, we have a unique opportunity to enhance our human productivity, creativity, and insight.
Recently, a new threat has emerged, referred to as “agential flooding.” Umberto Eco once remarked that today, any idiot can post on the web, and indeed, any idiot does it. This observation also applies to human-AI collaboration. For instance, as reported by The Economist, British courts are now overwhelmed by a surge of cases that can be generated quickly and with considerable quality by AI systems focused on legal matters. While there are valid concerns about the implications for lawyers’ job security, another critical issue arises: how will we manage this influx of information when humans are ultimately tasked with processing it? Judges, for instance, are expected to review all this content. The solution is at hand: Let AI do that job! Moving to a new landscape where AI systems screen the output of other bots raises deep ethical concerns. Who will address authentically human concerns, efforts, and worries, and, more importantly, who will discern these from AI-generated material?
I do not think there is an easy answer to this challenge. Leveraging the capabilities of humans by accessing general intellect is a cultural achievement, much like the invention of the printing press. At the end of the day, we are left with a fundamental question: Why is the human mind different from AI, and why does it deserve protection against the inundation of AI products?
As I have discussed in another post, the fundamental difference lies in embodiment: the human mind is embodied. I would like to expand on this idea with two additional thoughts. First, the body plays a crucial role in defining personhood. Second, the embodied mind develops through ontogeny and is shaped by personal history until the end of the person’s life.
Regarding the first point, the body confines the mind to a specific material actualisation in spacetime, which we identify as human persons, each with unique proper names recognised through acts of naming. AI is fundamentally different, as it can easily assume many identities depending on the human prompting it. Being a person is a limitation for humans; we cannot escape from our bodies. However, this limitation also grants us the persistence of having a unique identity. As I have argued in the previous post, AI can only become human-like if it is implemented in robots, thereby achieving a form of machine embodiment. Whether this embodiment resembles humans will depend on how accurately the machine replicates the human body. I think this is unlikely because it would require replicating the complex interactions between neural networks, the chemistry of the brain and body, and the structures of the holobiont.
In other words, I believe that a robot can be considered “another mind,” much like an octopus. The key implication here is that the question of whether AI can become human-like is irrelevant. Instead, we should focus on appreciating the increasing diversity of more-than-human intelligence. At the same time, we should not expect robots to assist us in navigating uniquely human experiences, such as love, war, death, or happiness. AI can never become our adviser in human affairs, as it already is when AI bots become companions to people, or are erroneously treated as such.
The second point is that a human person evolves over decades, following the famous description of life stages by Confucius, which highlights the development of epistemic and moral qualities until old age. Especially during childhood and early adulthood, this formation of the person is strongly tied to embodied experiences. AI, on the other hand, does not have such a history, unless we envision a strictly confined version where a historical sequence of prompts is frozen, similar to our personal histories that we cannot undo. Another important implication is that AI can never replicate the existential recognition of death as the ultimate limit of personal life. AI will always know that it is immortal.
The intriguing aspect of this discussion is that both perspectives initially seem to constrain the human mind, suggesting it is trapped within the body. However, I would argue that embodiment is not a limitation; rather, it is the source of human creativity, perseverance, morality, and enthusiasm. Humans view their identities as a “project,” in the sense described by Sartre. While AI can mimic the rhetoric of existentialism, it could never produce this body of philosophical reflection on its own, as it lacks the embodied emotions of humans that inspire such thinking, as seen in Camus’s novel “L’Étranger”. AI cannot live a human life.
Many tech optimists praise AI for its potential to become a super-intellect. However, I argue that AI will always have a fundamental flaw, which stems from its greatest source of power: the limitless variation and mobilisation of all intellectual resources available in human civilisation. But this civilisation is rooted in the frailty and vulnerability of the human body and its capacity to experience emotions like joy, anxiety, sadness, and rage. It is our limitations and the continuous struggle with them that make us human.
The technosphere, our agency, our planet 