Course Identification

History and Philosophy, Intelligence and Computing

This is a interdisciplinary course showing the relevance of the humanities (philosophy, history, religious studies) to cutting-edge current technologies, such as Artificial Intelligence. This relevance is not peripheral (like Ethical issues are) but addresses the core concern of developing novel algorithms.

This course presents a history of ideas relating to computing and Artificial Intelligence, beginning with the pre-Socratic philosophers, going through an overview of existing AI, and ending with blueprints for future programming, with videos of example runs.

Opening:

1. Introduction – Why History and Philosophy? Introduction to the two fields using the story of Positivism form the French revolution, via Auguste Comte to current secular thought.
2. Overview, Dreams, Background – Overview – the interaction between 3 carriers of intelligence. Dreams - The fascination with thinking machines from ancient theologies through medieval Islam into some failed attempts of the 19^th and early 20^th century. Background: Pan-computational theories, some logic, mathematics, algorithms.

Cultural Background:

3. Rationalism – About “isms”. The temptation to over-simplify. The irrational belief that rationality (Logic, Mathematics) will solve all problems. Predicate calculus.
4. Science and Secularization – The rise of modern critical thinking (Protestantism). The canonical tale of the evolution of modern astronomy. The myth of a “scientific clean sweep” (Descartes and Bacon). The rise of science up to scientific psychology.
5. The Modern Mindset - Kant and the Analytic-Continental split. Recap reformation and enlightenment. Romanticism and the Victorian era. Recap of Positivism and WW1. Modernism: worshipping efficiency, from architecture to agriculture. The rise of the MBA way of thinking (Simon).

The rise of AI

3. Pre-AI fields – Mathematics: state automata, boolean logic. Cybernetics: Tortoises, Punched cards, telegraph. Psychology and Linguistics: Behaviourism, review of Cognitive psychology and Linguistics. Electrical engineering: oscilloscopes and telephone exchanges. Biology – The discovery of neural networks (1943).
4. USA in the 1940-50s - birth of computation and AI – Freedom of will (Weber & Andersen). WW2: America’s self image as sanity between the Nazis and Communists, Vietnam War and the Cold-war philosophy. Von Neumann and the first computers. Cognitive revolution, founding AI: Dartmouth conference of 1956.

Overview of AI:

8. GOFAI – Good, Old-Fashioned, Artificial Intelligence – Definitions of Intelligence and AI. Logic, Fuzz, and experts. IBM, open-source. Herbert Simon’s pivotal role. Checkers, Chess, Go. Eliza. Winograd’s SHRDLU. The Nuclear bomb in the background of it all.
9. Turing Test and problems with knowledge – Turing (1950). Variants and criticisms. Was it a joke? “Common sense” etc. The fear of X taking over from humanity. CYC. Knowledge that vs knowledge how. Atomism and rationalism. Knowledge as true, justified belief and the infinite regress of “justified justifications”.
10. Biologically-inspired AI – Passive walking robots. Genetic Algorithms. Neural nets. Deep learning / Big Data. Simulating insects (Brooks). Histomimetic robotics – Hockings.

Critics of AI

11. Phenomenology – Philosophical background: phenomenology and existentialism. Four broken promises of AI, four assumptions necessary for AI, four distinctions, skills, sweeping pessimism. A fateful game of Chess. First-step falacy.
12. Other Critiques – Replies to the above critique. Intelligence embedded and embodied. Winograd & Flores (1986). Visions of the future: Singularity, trans-humanism. Optimism and Pessimism. Fiction inspiring technology.

Future Directions and Conclusion

13. Post-rationalist AI – Technology vs Science. Human-like vs Rational AI. Papert. Anthropic AI. Subjectivity and Introspection. Legitimacy issues. A-priori causes for optimism. Steps for developing introspection-based algorithms. Examples + implementation videos.
14. Ethics and Conclusion – Ethics and Technophobia: Technology as a threat to workers, and to religions. Technology changing human thought. The future arriving prematurely. Current social problems: leaks, social control, perpetuation of prejudices (big data). AI safely. Conclusion – a synchronous review of the strands discussed earlier.

Upon successful completion of this course students will:

1. Become familiar with key issues in the history & philosophy of AI;
2. Will develop an appreciation of the historical accidents that lead to our current mindset;
3. Develop abilities to raise philosophical questions about AI

TBD