Philosophy of the Machines: A Manifesto for Humans in the Age of Artificial Agents

We must reconsider our relationship with machines as artificial intelligence (AI) evolves from task-based support to autonomous or assisted generation. Generation is not creation. Human intelligence—our only known model —is the sole benchmark for evaluating AI. Yet, we lack rigorous comparison standards because we do not fully understand the internal mechanisms of human intelligence. Perhaps it is more appropriate to speak of AI in simplified terms—as something that, under certain conditions, emulates what we commonly perceive as intelligent behavior. But even that remains uncertain. Let us remember that we still lack a universally accepted definition of intelligence, let alone a definition of thinking, or even a deeper understanding of the complex nature of consciousness. Let us also remember that over the last 80 years of computing and automation, the dominant discipline for solving problems has not been AI, but engineering. Engineering provides the methodologies, mental models, and validation frameworks we use to design and deploy systems. AI extends this legacy—and at the same time disrupts it, introducing new epistemic and ethical challenges that cannot be resolved solely through efficiency. On these premises, this Manifesto articulates the Philosophy of the Machines as a distinct discipline that consolidates established philosophical lines of inquiry into a unified, applicability-oriented corpus with an explicit order of inquiry. Presented as a manifesto, it develops this order across ten interdependent sections. The trajectory culminates in the Δ–η–ζ framework, introduced as a modelling and analytical scaffold—and, where operationally feasible, a basis for measurement—of net gain in real deployments. The aim is to support more realistic, auditable, and human-aligned business cases for AI and generative AI in a labor landscape that must evolve as artificial agents become pervasive and socio-technical complexity continues to expand. Within this order of inquiry—moving from foundational analysis to an operational modelling scaffold—the Manifesto foregrounds three questions: (i) What kind of “intelligence” are we building? (ii) What kind of humans must we become in response? (iii) How should responsibility be allocated when systems exceed our capacity for full understanding and oversight? These questions set the Manifesto’s agenda for the age of artificial agents.