The shelf behind the course.

The thinking the course is built on — bias, statistics, verification, and computational thinking — each mapped to the sessions it feeds.

• May Contain Lies

  2024

  Alex Edmans

  Source of the Ladder of Misinference (statement → fact → data → evidence → proof) and the line that smart people are better at biased search; the course-wide answer-decoder. A hallucinated citation is a statement costumed as binding proof. Sessions 1–2.

• Thinking, Fast and Slow

  2011

  Daniel Kahneman

  System 1 and System 2; anchoring (including the sentencing-dice study on judges) and confirmation bias, mapped onto both legal judgment and LLM failure modes. Sessions 1–2.

• Noise

  2021

  Kahneman, Sibony & Sunstein

  On noise in judgment (identical cases, different sentences) applied to judges — and to a model's run-to-run variability. Sessions 1–2.

• Weapons of Math Destruction

  2016

  Cathy O'Neil

  Opaque, biased algorithmic decision systems in criminal justice, hiring and credit. Session 5.

• Hello World

  2018

  Hannah Fry

  Demystifies algorithms and where they go wrong; source of the centaur idea — human judgment paired with machine speed. Sessions 5 and 8.

• You Look Like a Thing and I Love You

  2019

  Janelle Shane

  Accessible, funny on-ramp to how AI fails (giraffing; recipes calling for broken glass; the tank/sunny-day shortcut). Session 5.

• The Art of Statistics

  2019

  David Spiegelhalter

  Base rates and the Harold Shipman detection; the statistics behind expert evidence. Session 2.

• How to Lie with Statistics

  1954

  Darrell Huff

  Statistical-misuse source; teach the credibility irony (Huff's tobacco work). Background reading; not covered in a core session.

• Calling Bullshit

  2020

  Bergstrom & West

  Verification heuristics: who's telling me this, how do they know, what are they selling. Sessions 6 and 8.

• Science Fictions

  2020

  Stuart Ritchie

  On research-integrity failures; supports the credibility-irony teaching (e.g. the Ariely fabrication scandal). Background reading; not covered in a core session.

• Computational Thinking, CACM 49(3)

  2006

  Jeannette Wing

  Foundational CT source; quote (under 15 words): computational thinking is a fundamental skill for everyone. Sessions 3–4.

• Mindstorms

  1980

  Seymour Papert

  Computational-thinking foundations. Sessions 3–4.

• Algorithms to Live By

  2016

  Christian & Griffiths

  Source of the 37% / explore-exploit idea for prompt iteration discipline. Session 4.

• The Pattern on the Stone

  1998

  W. Daniel Hillis

  A plain-language account of what computers actually do — the ground under 'thinking like a computer'. Session 3.

• “Playing Dice with Criminal Sentences” (Personality and Social Psychology Bulletin)

  2006

  Englich, Mussweiler & Strack

  The sentencing-dice study described in Thinking, Fast and Slow: judges who rolled a higher number handed down longer sentences. Session 1.

• Tomorrow's Lawyers

  Richard Susskind

  The changing legal job and professional identity — routine work automated, judgment and verification the premium skills. Sessions 7–8.

• Online Courts and the Future of Justice

  Richard Susskind

  Future of justice and legal work; access to justice. Background reading; not covered in a core session.

• The Black Box Society (2015)

  Frank Pasquale

  Opacity of algorithmic systems. Background reading; not covered in a core session.

• New Laws of Robotics (2020)

  Frank Pasquale

  Regulating AI; the human in the loop. Background reading; not covered in a core session.

• Artificial Intelligence and Legal Analytics (2017)

  Kevin D. Ashley

  What machines can and cannot do with legal texts — the ground under Session 7's map of reliable vs. flattened summarisation.

• Is Law Computable? (2020)

  Deakin & Markou (eds.)

  On the computability of law; background law-and-technology reading.

• “Deep Fakes: A Looming Challenge for Privacy, Democracy, and National Security”, 107 California Law Review 1753 (2019)

  Chesney & Citron

  The legal analysis of synthetic media and the source of the “liar's dividend”. Session 6.

1. 01
   Mata v. Avianca, Inc.678 F. Supp. 3d 443 (S.D.N.Y. 2023)

   Judge Castel; $5,000 sanction; six fabricated cases (Varghese, Martinez, Shaboon, Petersen, Miller, Estate of Durden); the lawyer's fatal assumption that ChatGPT could not possibly be fabricating cases. The course's flagship cautionary case (Sessions 1 and 8).

2. 02
   Gummadi Usha Rani v. Sure Mallikarjuna RaoSLP (C) No. 7575 of 2026 (SC of India, Narasimha & Aradhe JJ.)

   A pending Special Leave Petition in which the Supreme Court of India, taking note of a trial court order built on fake AI-generated judgments, observed that such a decision “would be a misconduct and legal consequence shall follow” and issued notice (amicus appointed). Not a final holding; the India-first anchor for the duty to verify (Sessions 1 and 8).

3. 03
   Deepak v. Heart & Soul Entertainment Ltd.Bombay HC, 7 Jan 2026 (Sathaye J.)

   ₹50,000 cost imposed for “dumping” unverified AI-generated written submissions citing a non-existent judgment; a concrete Indian consequence for unverified AI citations (Sessions 1 and 8).

4. 04
   Buckeye Trust v. PCITITA No. 1051/Bang/2024 (ITAT Bengaluru, 2024–25)

   Bengaluru ITAT order recalled under s.254(2) after it relied on ChatGPT-fabricated, non-existent case citations; the verification failure made concrete in an Indian tribunal (Session 8).

5. 05
   State v. Loomis881 N.W.2d 749 (Wis. 2016)

   Risk-assessment (COMPAS) and due process; the Wisconsin Supreme Court upheld use of a proprietary risk score in sentencing, with limits. Algorithmic bias in criminal justice (Session 5). Comparative (US).

6. 06
   Da Silva Moore v. Publicis Groupe287 F.R.D. 182 (S.D.N.Y. 2012) (Peck M.J.)

   First judicial approval of predictive coding / Technology-Assisted Review in e-discovery — AI deployed transparently and defensibly (Session 8). Comparative (US).

7. 07
   R v Sally ClarkUK, conviction quashed 2003

   Wrongful conviction driven by a statistical fallacy in expert evidence (the 1 in 73 million error); the definitive law-meets-statistics cautionary tale and the prosecutor's fallacy (Session 2).

8. 08
   NYT v. OpenAI / MicrosoftS.D.N.Y., No. 1:23-cv-11195 (pending)

   IP/copyright in training data and AI output; key claims survived a motion to dismiss. Comparative (US); status may have moved. Background reading; not covered in a core session.

9. 09
   ANI Media v. OpenAIDelhi HC, CS(COMM) 1028/2024 (Bansal J.; order reserved)

   India's first generative-AI copyright suit over training data; interim order reserved after ~32 hearings. Status may have moved. Background reading; not covered in a core session.

10. 10
    ProPublica, “Machine Bias” (COMPAS investigation)ProPublica, 23 May 2016

    Angwin et al.'s investigation finding racial disparities in the COMPAS recidivism risk tool; the empirical backbone of the algorithmic-bias discussion (an investigative report, not a case). Session 5. Comparative (US).

11. 11
    Jaswinder Singh v. State of PunjabP&H HC, 2023 (Chitkara J.)

    Punjab & Haryana High Court consulted ChatGPT for a “broader picture” of bail jurisprudence (it did not decide the bail on the AI output); a real Indian example of AI as navigation, not authority (Session 7).

12. 12
    Amazon's abandoned AI recruiting tool (Reuters report)Reuters, 10 Oct 2018 — a news report, not a judgment

    The well-known hiring case: a recruiting model trained on ten years of past CVs learned to penalise applications that mentioned women, and was scrapped. Session 5's real-world case of a biased predictive system in hiring. Verify to source before teaching.

• DPDP Act 2023

  India's Digital Personal Data Protection Act 2023 (enacted 11 Aug 2023); the DPDP Rules 2025 were notified Nov 2025 and are phasing in to 2027. Grounds the confidentiality discipline — no personal or client data into public LLMs (Session 8).

• EU AI Act

  Regulation (EU) 2024/1689; entered into force Aug 2024 and phasing in to 2027. Comparative regulatory background; not covered in a core session.

• Supreme Court of India White Paper on AI and the Judiciary

  Centre for Research and Planning, Supreme Court of India, Nov 2025. Warns of hallucinations, bias, and confidentiality; stresses mandatory human verification and the judge as ultimate decision-maker; restricts cloud GenAI for case data. The India-first frame for Sessions 7–8.

From the generalist chatbots you already know to the judiciary's own approved systems — surveyed so you can judge fit, not endorse a vendor.