A Developmental History of Artificial Intelligence

About the Book

Where preparation became a science

Volume II picks up exactly where Volume I ends — at the moment six millennia of preparation became a working science. It begins with McCulloch and Pitts's founding neuron model in 1943 and closes at the audited snapshot of the Stanford AI Index 2026, tracing the nonlinear, cyclical rise of AI across six eras and continuing Volume I's chapter numbering.

• The Birth of the Field (1943–1956) — from the founding papers to the Dartmouth naming
• Symbolic AI & the First Winter (1956–1973)
• Expert Systems & the Second Winter (1974–1993)
• Neural Revival & Deep Learning Ascendance (1994–2016)
• The Transformer Revolution & Foundation Models (2017–2022)
• The Agentic & Multimodal Era (2023–2026)

Built on a citation-verified archive — 85 annotated entries with APA sources, a tiered bibliography of more than 180 works, and the complete AI Top-5 Influencers Timeline (420 ranked entries, 1943–2026) as a decade-by-decade appendix — the book shows how each paradigm emerged, matured, collapsed, and re-formed into something stronger: Hebb's learning rule and Rosenblatt's perceptron through backpropagation, convolutional networks, attention, the transformer, and the reasoning and agentic systems of 2024–2026.

The threads of Volume I continue forward, now six: formal logic into reasoning models; mechanical calculation into GPU superclusters and agents; the theological thread from the Golem to Constitutional AI and the EU AI Act; behavioral learning from Thorndike to RLHF; the mathematics of uncertainty from Bayes to the scaling laws; and knowledge organization from ontologies to retrieval-augmented generation.

The field's recurring patterns — the hype–disillusionment pendulum, dead ideas reborn, the Bitter Lesson — documented across eight decades.