Dissertation Abstract: This dissertation asks what happens to human nature when the environment that shaped it begins to change rapidly and in ways that are difficult to foresee. Its point of departure is that artificial intelligence is more adequately understood not as a tool but as an environment: once AI systems become part of the infrastructure of communication, decision-making, education, and administration, they alter the criteria of success for human action and institutional design, and thereby act on human practices in a way that can be analyzed through the concept of selection pressure. The project uses this concept non-metaphorically. It develops a methodological apparatus that distinguishes genuine selection from optimization and from other stabilizing mechanisms, and it formulates diagnostic questions for deciding when selectionist language is warranted and when it is merely rhetorical. On this basis it examines four philosophical conceptions of human nature, those of Hobbes, Kant, Dennett, and Sterelny, across the divide between essentialist and anti-essentialist views. Its central claim is that, despite their deep differences, all four make the realization of human nature dependent on environmental conditions, so that the AI environment is philosophically consequential whichever conception one holds. The dissertation is deliberately neutral on the essentialism dispute: its structural conclusions do not depend on taking a side.
The work proceeds in four chapters. The first lays the conceptual and methodological foundations and argues that the AI environment is selection-relevant, identifying four features not jointly present in earlier sociotechnical formations and analyzing the shift of institutional practice from goals to measurable proxies. The second examines the competitive development of AI as a strategic environment of Hobbesian structure, in which states and developers act under mutual uncertainty without a superior authority to enforce agreements. The third turns to autonomy and authenticity, treating AI both as a threat to the conditions of their exercise and as a provocation to self-overcoming, so that authenticity becomes a task rather than a given. The fourth extends the analysis to environments shaped by advanced AI systems (AGI, ASI) and asks which conditions of human stability would be preserved, transformed, or endangered there.