LPS 221: Knowing How, Knowing That, and AI

Presentation of the motivating idea for the class (slides), discussion of logistics, and sign-ups for dates and topics.

Presented by Neil.

Required

• Videos 1, 2, 3, and 5 of the 3blue1brown sequence on neural networks. It’s fine if you don’t follow all the math in part 2; if you do, watch 4 as well.
• Timothy B. Lee, “How computers got shockingly good at recognizing images”, Ars Technica (December 2018)
• Timothy B. Lee and Sean Trott, “A jargon-free explanation of how AI large language models work”, Ars Technica (July 2023)
• At least skim chapter 1 of Russell and Norvig, Artificial Intelligence: A Modern Approach (2010)

Further resources — neural nets

• Neural network playground — try out the basics on simple datasets
• The rest of the 3blue1brown sequence
• Michael Nielsen, Neural Networks and Deep Learning — a free online book that teaches you to build one
• Emergent Garden, “Why neural networks can learn (almost) anything” (March 2022)
• Art of the Problem, “How neural networks learn — backpropagation intuition” (November 2019)
• Wikipedia, “History of artificial neural networks”

Further resources — large language models

• Vaswani et al., “Attention is all you need” (2017) — the paper that launched large language models
• Vox, “Computers just got a lot better at writing” (March 2020) — a video about GPT-3; notice the date
• Textsynth — try out pure predict-the-next-word LLMs. They work better given paragraphs as a prompt, but will attempt part of a sentence.
• Ben Levinstein’s five-part series explaining large language models (January–February 2023)

Further resources — interpretability and robotics

• Welch Labs, “The moment we stopped understanding AI” (July 2024)
• Golden Gate Claude (May 2024) — a step forward in interpretability
• Brian Potter, “Robot dexterity still seems hard” (April 2025)

Presented by Andrew.

Required

• Gilbert Ryle, “Knowing how and knowing that: the presidential address” (1946)
• Jason Stanley and Timothy Williamson, “Knowing How” (2001)

Supplementary

• Gilbert Ryle, The Concept of Mind, chapter 2, “Knowing how and knowing that” (1949)
• Barry Smith, “Knowing how vs. knowing that” (1988)
• Lewis Carroll, “What the tortoise said to Achilles” (1895)
• Jason Stanley, Know How, Oxford University Press (2011) — with reviews by Kent Bach and Robert Stalnaker
• Stephen Hetherington, How to Know: A Practicalist Conception of Knowledge, Wiley (2011) — with a review by B. J. C. Madison
• Ephraim Glick, “Practical modes of presentation” (2015)
• Alexander Kocurek and Ethan Jerzak, “Knowing what to do” (2024)

Presented by Banin and Yunlong.

Required

• Jason Stanley and John Krakauer, “Motor skill depends on knowledge of facts” (2013)

Discussed by the presenters

• Neil Levy, “Embodied savoir-faire: knowledge-how requires motor representations” (2017)
• Ellen Fridland, “Skill and motor control: intelligence all the way down” (2017)
• David Papineau, “In the Zone” (2013)

Possibly relevant

• Alexander Mugar Klein, Consciousness is Motor (2025)
• Brian Potter, “Robot dexterity still seems hard” (April 2025)
• Rodney Brooks, “Why today’s humanoids won’t learn dexterity” (September 2025)

Presented by Lauren and Elijah.

Required

• Alvin Goldman, “Discrimination and perceptual knowledge” (1976)
• Ernest Sosa, “How to defeat opposition to Moore” (1999)
• J. Adam Carter and Duncan Pritchard, “Knowledge-how and epistemic luck” (2015)

On modal properties of knowledge

• Ernest Sosa, “How must knowledge be modally related to what is known?” (1999)
• Alvin Goldman and Bob Beddor, “Reliabilism”, Stanford Encyclopedia of Philosophy (2021)

On knowing how and luck

• Katherine Hawley, “Success and knowledge-how” (2003)
• Carlotta Pavese, “Know-how, action, and luck” (2018)

Presented by Tori and Tessa.

Required

• Alan Turing, “Computing Machinery and Intelligence” (1950)
• Hubert Dreyfus, “From Socrates to expert systems: the limits of calculative rationality” (1987)

Supplementary

• Stuart Russell and Peter Norvig, Artificial Intelligence: A Modern Approach, chapter 26
• Richard Ngo, “Towards a scale-free theory of intelligent agency” (2025)
• G. W. F. Hegel, “Who thinks abstractly?” (1807)
• Hubert Dreyfus and Stuart Dreyfus, “What artificial experts can and cannot do” (1991)
• Hubert Dreyfus, “Overcoming the myth of the mental” (2005)
• Hubert Dreyfus, “Why Heideggerian AI failed and how fixing it would require making it more Heideggerian” (2007)

Presented by Mandy.

Required

• Stuart Dreyfus and Hubert Dreyfus, “A five-stage model of the mental activities involved in directed skill acquisition” (1980)

Supplementary

• Stuart Dreyfus, “System 0: the overlooked explanation of expert intuition” (2014)
• Adolfo Peña, “The Dreyfus model of clinical problem-solving skills acquisition: a critical perspective” (2010)
• PBS video of Daniel Dennett and Hubert Dreyfus on Garry Kasparov’s loss to Deep Blue in 1997

Presented by Chris and Ryan.

Required

• Jonathan St. B. T. Evans, “In two minds: dual-process accounts of reasoning” (2003)
• Jonathan St. B. T. Evans and Keith Stanovich, “Dual-process theories of higher cognition: advancing the debate” (2013)
• Wim de Neys, “Advancing theorizing about fast-and-slow thinking” (2022) — read the main article, maybe glance at the responses

Supplementary

• Steven Sloman, “The empirical case for two systems of reasoning” (1996)
• Gideon Keren and Yaacov Schul, “Two is not always better than one: a critical evaluation of two-system theories” (2009)

Presented by Julia and Jaehyun.

Required

• Katherine Hawley, “Testimony and knowing how” (2010)
• Thi Nguyen, “Transparency is surveillance” (2021)

Further reading

• Katherine Hawley, “Success and knowledge-how” (2003)
• Joshua Habgood-Coote, “What’s the point of knowing how?” (2019)
• Jennifer Lackey, “Knowing from testimony” (2006)
• Annette Baier, “Trust and antitrust” (1986)

Presented by Yifan.

Required

• Jane X. Wang, “Meta-learning in natural and artificial intelligence” (2021)
• Nicole Drummond and Yael Niv, “Model-based decision making and model-free learning” (2020)
• Kate Nussenbaum and Catherine Hartley, “Understanding the development of reward learning through the lens of meta-learning” (2024)

Further reading

• Machiel Keestra, “Metacognition and reflection by interdisciplinary experts” (2017)
• Stephen Fleming and Hakwan Lau, “How to measure metacognition” (2014)
• Bernard Baars, “Global workspace theory of consciousness” (2005)
• Simon Goldstein and Ben Levinstein, “Does ChatGPT have a mind?” (2024)

Additional readings on skill and know-how

• Ellen Fridland, “Learning our way to intelligence: reflections on Dennett and appropriateness” (2015)
• Carlotta Pavese, “Practical knowledge first” (2022)
• Yuri Cath, “Knowing how without knowing that” (2011)
• Ellen Fridland, “They’ve lost control: reflections on skill” (2014)
• Carlotta Pavese, “Skill in epistemology I: skill and knowledge” and “Skill in epistemology II: skill and know how” (2016)
• Jason Stanley and Timothy Williamson, “Skill” (2017)

Additional readings on meaning and LLMs

• Emily Bender and Alexander Koller, “Climbing towards NLU: on meaning, form, and understanding in the age of big data” (2020)
• Harvey Lederman and Kyle Mahowald, “Are language models more like libraries or like librarians?” (2024)
• Matthew Mandelkern and Tal Linzen, “Do language models’ words refer?” (2024)