Readings
Readings are mainly available on Perusall, but are also linked here.
Week 1
- Design Norms
- Animation: Explaining Technological Mediation - YouTube - Technology shapes us.
- How to read a paper — Keshav, 2007 - suggested reading/skimming time: 20 min
- Jobs-to-be-done Framework
Week 2
This week’s readings focus on technical background, history, perspectives, and possible futures.
- Large Language Models explained briefly - YouTube (8 minutes): An insightful big-picture summary of how LLMs work. Feel free to speed it up if you already have some background.
- Three AI Futures: This article in the most recent Communications of the ACM caught my eye because it gives a few different perspectives about AI. Each viewpoint is a caricature that’s missing some nuance, so feel free to disagree.
- Wikipedia article on the classic Licklider 1960 paper, Man-Computer Symbiosis, written at another moment of AI enthusiasm. What aged well? (I’ve included the full original paper in the Perusall library if anyone wants to read it, there’s some good quotes.)
- Welcome to the Era of Experience: this recent paper has helped shape my thinking recently: with scaling for mimicry tasks hitting a wall, the next major move will be learning from interaction. I’ll probably assign some related readings later: The Second Half – Shunyu Yao – 姚顺雨, The Era of Exploration | Yiding’s blog
If you haven’t yet, I also recommend the AI Fluency Foundations course; comment on that in the discussion forum.
Week 3
- Guidelines for Human-AI Interaction | CHI 19 Amershi et al., 2019
suggested reading/skimming time: 1 hr - Enough AI copilots! We need AI HUDs - “[spell check] just instantly adds red squigglies when you misspell something! You now have a new sense you didn’t have before. It’s a HUD [heads-up display].”; “routine predictable work might make sense to delegate to a copilot, but for extraordinary outcomes, … equip human experts with new superpowers.”
- How to Fix Your Context
- Why Our Small Business Chooses Human Intelligence Over AI - NeoMam Blog
Week 4
This week’s readings juxtapose three perspectives on AI’s impact on work:
- AI-Generated “Workslop” Is Destroying Productivity
- 2025: The Year the Frontier Firm Is Born
- Homogenization Effects of Large Language Models on Human Creative Ideation C&C 2024
And two perspectives on AI’s impact on society:
- Antiqua et nova. Note on the Relationship Between Artificial Intelligence and Human Intelligence (28 January 2025)
- Will AI kill everyone? Here’s why Eliezer Yudkowsky thinks so. | Vox
In class I also referenced the following (optional):
- Thinking, Searching, and Acting - by Nathan Lambert
- Various Inference Providers - I showed Together AI for one example.
- You may also be interested in OpenRouter for LLM API flexibility.
Reading 5
- AI Idolatry and Magical Thinking – Ken Arnold (Calvin U)
- Andon Labs Safety Report, August 2025 - How do LLMs do at entrepreneurship? spoiler alert: badly.
- The Leader’s Playbook for GenAI Metrics
- The Bubble That Knows It’s a Bubble - “The most radical act might be patience.”, “Can you remain skeptical during euphoria and optimistic during despair?”, “We’ve seen this before.”
- A Good Plan is Hard to Find - “aligning helpful LLMs needs feedback from real user interactions—not just preferences of what looks helpful”
- AI, layoffs, productivity and The Klarna Effect - “After firing 700 humans for AI, Klarna now wants them back”; “The talking point that LLMs will make workers 10x more productive is probably not accurate.”
- The Slow Collapse of Critical Thinking in OSINT due to AI
Additional optional readings
Which of these readings should we incorporate in a future version of this course?
- Building software on top of Large Language Models
- When Our Kid Has a Human and an AI Lover: A Conversation with Alexandra Diening on the Future of Relationships | ACM Interactions
- Calm Down—Your Phone Isn’t Listening to Your Conversations. It’s Just Tracking Everything You Type, Every App You Use, Every Website You Visit, and Everywhere You Go in the Physical World - McSweeney’s Internet Tendency
- “Human-Centered Artificial Intelligence” by Ben Shneiderman
- The lethal trifecta for AI agents: private data, untrusted content, and external communication
- AI as Normal Technology | Knight First Amendment Institute
- Anti-Heroes
- Creating Value for Other People - YouTube
- Deskilling and Healthcare AI - by Robert Wachter
- LLM Engineer Roadmap
- The following readings are taken from Coaugmentation.com
- Augmenting human intellect: A conceptual framework – Engelbart, 1962
suggested reading/skimming time: 1 hr, 15 min - Designing for human–AI complementarity in K-12 education — Holstein & Aleven, 2021
suggested reading/skimming time: 45 min - Cognitive Science of Augmented Intelligence - Dubova - 2022 - Cognitive Science - Wiley Online Library
- Towards human-AI deliberation: Design and evaluation of LLM-empowered deliberative AI for AI-assisted decision-making — Ma et al., 2024
suggested reading/skimming time: 45 min - Trust in automation: Designing for appropriate reliance — Lee & See, 2004
suggested reading/skimming time: 1 hr, 15 min - Improving student learning with hybrid human-AI tutoring: A three-study quasi-experimental investigation – Thomas et al., 2024
suggested reading/skimming time: 45 min - Persistent Assistant: Seamless everyday AI interactions via intent grounding and multimodal feedback – Cho et al., 2025
suggested reading/skimming time: 1 hr - Exploring the potential for generative AI-based conversational cues for real-time collaborative ideation – Rayan et al., 2024
suggested reading/skimming time: 45 min - The ideation–execution gap: Execution outcomes of LLM-generated versus human research ideas – Si, Hashimoto, & Yang, 2025
suggested reading/skimming time: 45 min - Productive vs. reflective: How different ways of integrating AI into design workflows affect cognition and motivation – Xu et al., 2025
suggested reading/skimming time: 1 hr - How large language models can reshape collective intelligence – Burton et al., 2024
suggested reading/skimming time: 1 hr - Judgment Sieve: Reducing uncertainty in group judgments through interventions targeting ambiguity versus disagreement Chen & Zhang, 2023
suggested reading/skimming time: 1 hr - AI can help humans find common ground in democratic deliberation | Science
- The metacognitive demands and opportunities of generative AI – Tankelevitch et al., 2024
suggested reading/skimming time: 1 hr
- Augmenting human intellect: A conceptual framework – Engelbart, 1962