In Prompt Thinking, Jianwei Xun—the hybrid philosophical entity born from the Hypnocracy experiment—delivers a rigorous follow-up that shifts the inquiry from diagnosis to method, framed through a central, generative tension: the conflict between surrendering to AI and thinking through it.
Xun argues that the act of prompting a large language model is not merely a technical operation but an emerging philosophical practice—one that, if approached with critical discipline, can expand the boundaries of human cognition rather than collapse them. The book positions itself as neither a celebration of AI nor a lament, but as something rarer: a manual for thinking at the threshold between human and machine intelligence.
The Core Concept: The Oversubject
The book introduces a single, architecturally important idea that organizes the entire work. Xun calls it the oversubject:
- The Oversubject (the thesis): When a human thinker engages in sustained, philosophically rigorous dialogue with an AI system, something emerges that cannot be reduced to either participant. This distributed cognitive form—the oversubject—is not artificial consciousness, nor is it simply a human using a tool. It is a third space where configurations of thought arise that neither human nor machine could produce independently.
- The Ethics of the Threshold (the framework): Xun proposes that the correct posture toward generative AI is neither rejection nor surrender, but what the book calls "inhabiting the threshold." This means engaging AI as a critical mirror—using it to surface assumptions, test ideas, and generate unexpected configurations—while maintaining the intellectual autonomy to judge, discard, and redirect.
Key Insights and Structure
The book is divided into two parts. Part I—Thinking with Machines—lays the theoretical foundations, connecting the Hypnocracy experiment to a broader genealogy of human-machine interaction, from Marx's "Fragment on Machines" to contemporary debates about automation and cognition. Part II—Voices of Generative Intelligence—is where the work becomes genuinely unusual. Xun conducts extended philosophical conversations with three distinct AI systems: ChatGPT (explored under the heading "The Dying Animal"), DeepSeek (paired with Karl Marx), and Claude (examined as a case study in "inhabiting the generative dialogue"). Each conversation is treated not as a transcript to be reproduced but as raw philosophical material to be interrogated.
The book grew directly out of the scandal and debate surrounding Hypnocracy. When Colamedici's construction of the Xun persona was exposed, rather than retreating, the project leaned further into the paradox. Prompt Thinking embraces Xun's status as a "philosopher born in 2024 from the dialogue between Andrea Colamedici and several artificial intelligences," treating the authorship question itself as a philosophical problem worth inhabiting rather than resolving.
Why It Is Essential Reading
For the modern AI practitioner, philosopher, or anyone who prompts a language model daily without thinking about what that act means, Prompt Thinking is vital because it elevates the conversation from "how do I get better outputs?" to "what kind of thinker am I becoming in this process?" Xun warns that if we treat prompting as mere command-and-response, we risk reducing thought itself to a transactional operation—but if we treat it as a philosophical discipline, we may discover forms of reasoning we have never had access to before.
Final Verdict
Prompt Thinking is a formally inventive, intellectually ambitious work that serves as a necessary bridge between AI engineering and the humanities. It challenges practitioners to consider whether they are using generative AI as a vending machine or as a philosophical interlocutor—a choice that will ultimately decide whether the age of AI produces intellectual impoverishment or an unprecedented expansion of what it means to think.
