Artificial Intelligence—Preliminary Notes
Juan Diaz-Granados · 2026
Chapter 2 lays the contextual and conceptual foundations for the discussion of AI and tort liability. It traces AI’s evolution from mid-20th-century symbolic, rule-based systems to machine learning’s data-driven paradigm and, more recently, to deep-learning architectures—especially “Transformers”—that power today’s large language models. The chapter then clarifies the conceptual framework adopted in this book, noting the absence of a single canonical definition and opting for a deliberately broad, legally oriented notion of AI as a technology capable of performing tasks associated with human intelligence and cognition, such as reasoning, learning and decision-making. Finally, the chapter provides a practical taxonomy of AI systems relevant to tort liability analysis, including machine learning, generative AI, autonomous AI, agentic AI and embodied AI—highlighting how increasing autonomy, planning and physical presence expand the spectrum of risks. This chapter presents three key preliminary explanations that provide the contextual and conceptual foundations for the discussion that follows. It offers a concise overview of the historical development of AI, clarifies the conceptual framework underpinning this book and distinguishes between the main types of AI relevant to the analysis. Understanding AI’s historical evolution, conceptual boundaries and technological distinctions is essential to engaging with the doctrinal and theoretical considerations explored later in the book.