Chapter 25 Further Reading: Decision Support, Analysis, and Strategic Thinking

Decision-Making and Judgment

"Thinking, Fast and Slow" Daniel Kahneman The foundational text on human judgment and cognitive biases. Kahneman's distinction between System 1 (fast, intuitive) and System 2 (slow, deliberate) thinking is the conceptual backbone for understanding when AI decision support helps (by systematizing System 2 analysis) and when it fails (when AI's fluent outputs feel like authoritative System 1 judgments). The planning fallacy, anchoring, and availability bias are all directly relevant to AI-assisted decision-making.

"Superforecasting: The Art and Science of Prediction" Philip Tetlock and Dan Gardner Research on what distinguishes good forecasters from poor ones — directly applicable to evaluating the quality of AI-generated predictions and scenario analyses. Tetlock's "foxes" (who draw on multiple perspectives) outperform "hedgehogs" (who know one big thing) — a pattern that maps to how AI decision support should be used: as one source of analysis among several, not as the single authoritative view. [goodjudgment.com]

"Thinking in Bets: Making Smarter Decisions When You Don't Have All the Facts" Annie Duke Duke's framework for probabilistic thinking under uncertainty is directly applicable to the scenario analysis and assumption testing techniques in this chapter. Her concept of "resulting" — judging decisions by outcomes rather than quality of reasoning — is the error that decision records are designed to prevent. [annieduke.com]

"Sources of Power: How People Make Decisions" Gary Klein Klein's research on naturalistic decision-making reveals how experts actually make decisions under pressure — through recognition-primed pattern matching, not rational analysis. This is important context for understanding where AI decision support adds value (in systematic analysis of novel situations) and where expert intuition remains superior (in time-pressured, familiar situations). Klein also originated the pre-mortem technique discussed in Chapter 24.

Strategic Analysis

"Good Strategy Bad Strategy: The Difference and Why It Matters" Richard Rumelt Rumelt's analysis of what distinguishes genuinely strategic thinking from "strategic" language that has no substance. His "kernel of strategy" (diagnosis, guiding policy, coherent actions) is a useful filter for evaluating AI-generated strategic analysis — does it identify a real diagnosis, or does it generate strategic-sounding language without analytical teeth?

"Competitive Strategy: Techniques for Analyzing Industries and Competitors" Michael E. Porter The canonical source for Porter's Five Forces and other competitive analysis frameworks that appear in this chapter. More detailed and nuanced than any AI prompt can capture. Worth reading if you use competitive analysis frameworks regularly and want to understand the full model rather than the textbook summary.

"Playing to Win: How Strategy Really Works" A.G. Lafley and Roger Martin Lafley and Martin's "strategy choice cascade" (winning aspiration, where to play, how to win, core capabilities, management systems) provides a structured framework for strategic decision-making that is well-suited to AI-assisted analysis. The book's emphasis on explicit choice — strategy as saying no — is a useful corrective to AI-generated strategic analyses that tend toward comprehensiveness over decisiveness.

Decision Frameworks and Tools

"Smart Choices: A Practical Guide to Making Better Decisions" John Hammond, Ralph Keeney, and Howard Raiffa A practical guide to structured decision-making that covers the same frameworks described in this chapter (objectives, alternatives, consequences, trade-offs, uncertainty) in greater depth. The PrOACT framework (Problem, Objectives, Alternatives, Consequences, Trade-offs) is a useful organizing structure for complex decisions.

"The Anatomy of a Decision" Harvard Business Review collection A curated collection of HBR articles on decision-making that covers cognitive biases, decision architecture, organizational decision processes, and decision quality. Useful as a reference collection rather than a linear read. Available through HBR.org or in print compilation.

"Decisive: How to Make Better Choices in Life and Work" Chip and Dan Heath The Heaths' WRAP process (Widen your options, Reality-test your assumptions, Attain distance, Prepare to be wrong) maps directly to the AI-assisted decision techniques in this chapter. Particularly good on assumption testing and the "10/10/10" temporal distancing technique for gaining perspective on decisions.

AI and Decision Quality

"Human Compatible: Artificial Intelligence and the Problem of Control" Stuart Russell Russell's analysis of AI alignment problems is relevant context for the "AI cannot weigh your values" point in this chapter. His framework for understanding what it would mean for AI to make decisions "for" humans provides a deeper grounding for why AI decision support must remain support — not delegation.

"The Alignment Problem: Machine Learning and Human Values" Brian Christian A more accessible exploration of how AI systems can produce outputs that are technically optimal but misaligned with what we actually want. Directly relevant to the "compelling but wrong" failure mode described in the chapter.

Ethics and Accountability

"A Rulebook for Arguments" Anthony Weston A brief, practical guide to constructing and evaluating arguments — directly relevant to the devil's advocate technique and the ability to assess whether AI-generated arguments are actually sound. Being able to identify logical fallacies, unsupported premises, and invalid inferences is an essential skill for evaluating AI decision analysis.

"Justice: What's the Right Thing to Do?" Michael Sandel Sandel's exploration of ethical frameworks (utilitarian, libertarian, egalitarian, communitarian) provides tools for thinking about values-laden decisions where AI analysis is most limited. Understanding the different ethical frameworks doesn't tell you what to decide, but it gives you vocabulary for the values trade-offs that AI cannot resolve for you. Based on his famous Harvard course, also available as a lecture series online at justice.harvard.edu.