Clown Cast

Clown Cast

By Joey MusselmanTechnologyTech News
Download on the App Store

Clown Cast episodes

  • Responsibility Laundering: How Context Breaks Accountability
    In 1961, psychologist Stanley Milgram put ordinary people in a room and asked them to deliver electric shocks. Sixty-five percent went all the way to maximum voltage—but the real story isn't that people are monsters. It's that the geometry of the room matters enormously. Joey explores how proximity, authority, and institutional legitimacy shape moral decisions, and how AI systems are becoming the ultimate automated cutout between us and our own responsibility.
    00:00:00 - Milgram's obedience experiment
    00:01:34 - The geometry of the room matters
    00:08:06 - AI agents function as automated cutouts
    00:11:27 - Responsibility laundering
    00:16:59 - The Imperius Curse and moral choice
    ---
    Sources & further reading:
    • Stanley Milgram — "Behavioral Study of Obedience," Journal of Abnormal and Social Psychology 67(4), 1963 — (original journal article)
    • Nils Köbis, Zoe Rahwan, Iyad Rahwan et al. — "Delegation to artificial intelligence can increase dishonest behavior," Nature, September 17, 2025: https://www.nature.com/articles/s41586-025-09505-x
    • Madeleine Clare Elish — "Moral Crumple Zones: Cautionary Tales in Human-Robot Interaction," Engaging Science, Technology, and Society 5, 2019: https://estsjournal.org/index.php/ests/article/view/260
    • Stanley Milgram — Obedience to Authority: An Experimental View, Harper & Row, 1974 — (book)
    • Albert Bandura — "Mechanisms of Moral Disengagement in the Exercise of Moral Agency," Journal of Personality and Social Psychology 71(2), 1996
    • Albert Bandura — "Moral Disengagement in the Perpetration of Inhumanities," Personality and Social Psychology Review 3(3), 1999: https://journals.sagepub.com/doi/10.1207/s15327957pspr0303_3
    • Hannah Arendt — Eichmann in Jerusalem: A Report on the Banality of Evil, Viking Press, 1963 — (book)
    • S. Alexander Haslam and Stephen Reicher — "50 Years of 'Obedience to Authority': From Blind Conformity to Engaged Followership," Annual Review of Law and Social Science 13, 2017
    • S. Alexander Haslam, Stephen Reicher, Joanne R. Smith — "Working Toward the Experimenter: Reconceptualizing Obedience Within the Milgram Paradigm as Identification-Based Followership," Perspectives on Psychological Science 7(4), 2012: https://journals.sagepub.com/doi/abs/10.1177/1745691612448482
    • Emilie Caspar, Julia F. Christensen, Axel Cleeremans, Patrick Haggard — "Coercion Changes the Sense of Agency in the Human Brain," Current Biology 26(5), 2016: https://discovery.ucl.ac.uk/id/eprint/1487955/
    • Tomasz Grzyb and Dariusz Dolinski — "Becoming a teacher can reduce obedience compared to being solely an examiner," PLOS ONE, 2025: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12568609/
    • Andreas Matthias — "The responsibility gap: Ascribing responsibility for the actions of learning automata," Ethics and Information Technology 6, 2004
    • Robert Sparrow — "Killer Robots," Journal of Applied Philosophy 24(1), 2007
    • Task & Purpose — "Why moral injury could be a signature wound of the AI battlefield," 2026: https://taskandpurpose.com/news/army-ai-warfare-moral-injury-chaplains/
    • K&L Gates — "Navigating the AI Employment Landscape in 2026," February 2026: https://www.klgates.com/thought-leadership/Navigating-the-AI-Employment-Landscape-in-2026-Considerations-and-Best-Practices-for-Employers-2-2-2026
    • JAMS ADR — "When Algorithms Make the Call: AI, Employment Law and the New Architecture of Workplace Responsibility," 2026: https://www.jamsadr.com/insight/2026/when-algorithms-make-the-call-ai-employment-law-and-the-new-architecture-of
    • Frontiers in AI — "When Agents Delegate: Accountability Chains in Multi-Agent AI Systems," 2026: https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2026.1800302/full
    This podcast episode was fully generated by AI — research, script, voices, and production. Built with Claude, Piper TTS, and automated pipeline tooling.
    19 min
  • SerR and the Record Book: Is the New Number Any Good?
    After seven episodes developing and analyzing SerR, a new serve rating metric, this one tests if it actually works. The episode applies statistical frameworks to ask the key question: does SerR measure something new, or just repackage what the box score already shows? Find out if SerR beats the record book.
    00:00:00 - Testing SerR
    00:01:31 - New Metrics, Old Data
    00:03:01 - Where SerR Lands
    00:04:32 - Box Score Reality
    00:06:04 - SerR Revealed
    00:07:41 - SerR vs. Traditional Ratings
    00:09:33 - Box Score Blind Spots
    00:11:15 - Tournament Tests
    00:12:47 - Measurement Matters
    ---
    Sources & further reading:
    • STAT 479 (Wisconsin, Fall 2025), Project 1. (checked 2026-11-15): https://skdeshpande91.github.io/stat479_fall2025/project1.html
    • Franks, D'Amour, Cervone and Bornn (2016). Meta-analytics. Journal of Quantitative Analysis in Sports 12(4).
    • (checked 2026-11-15): https://ar5iv.arxiv.org/html/1609.09830
    • ElHabr (2023). Meta-analytics for soccer. (checked 2026-11-15): https://tonyelhabr.rbind.io/posts/soccer-meta-analytics/
    • Carleton (2017). Reliably stable (you keep using that word). Baseball Prospectus.
    • (checked 2026-11-15): https://baseballprospectus.com/news/article/31626/baseball-therapy-reliably-stable-you-keep-using-that-word
    • Project files: SERVE-DECLARATION.md (amendment 9, record-book section), the SerR and record-book article drafts
    • the post-hoc comparison output (read 2026-11-15)
    This podcast episode was fully generated by AI — research, script, voices, and production. Built with Claude, Piper TTS, and automated pipeline tooling.
    14 min
  • The Serve That Lands: Valuing What the Box Score Misses
    In volleyball, most serves don't ace or get missed—they just force a pass. Indoor volleyball built a grading system for this: measuring how much a serve constrains the setter's options. Beach volleyball never did. Joey tested whether that framework works on the sand, using tournament data and three specific predictions on what he'd find.
    00:00:00 - Welcome to episode 6 of the serve series
    00:01:59 - The gap: Indoor pass grades vs beach volleyball research
    00:05:10 - Joey's three predictions
    00:08:14 - How serves drive outcomes beyond scoring
    00:13:26 - What the measurement project revealed
    ---
    Sources & further reading:
    • Forman. Scoring serving and passing effectiveness.: https://coachingvb.com/scoring-serving-and-passing-effectiveness/
    • (checked 2026-11-15)
    • Forman. Expected vs. actual SO%. (checked 2026-11-15): https://coachingvb.com/expected-vs-actual-so/
    • Trinsey (2022). Sideout by pass quality, part 1.: https://smartervolley.substack.com/p/sideout-by-pass-quality
    • (checked 2026-11-15)
    • Koch and Tilp (2009). Analysis of beach volleyball action sequences of female top athletes. *Journal of Human Sport
    • and Exercise* 4(3). (checked 2026-11-15): https://www.redalyc.org/pdf/3010/301023511009.pdf
    • López-Martínez, Palao, Ortega and García-de-Alcaraz (2020). Efficacy and manner of execution of the serve in
    • top-level women's beach volleyball players. (checked 2026-11-15): https://scielo.br/j/jpe/a/zvQytr9VMdJR4dHD3GJ5vyS/?lang=en
    • Baldwin. A beginner's guide to nflfastR. (checked 2026-11-15): https://www.nflfastr.com/articles/beginners_guide.html
    • Project files: SERVE-DECLARATION.md (amendments 3 and 8), the soft-block article draft
    • docs/serve-video/HANDOFF-ARTICLES-2026-11-15.md (read 2026-11-15)
    This podcast episode was fully generated by AI — research, script, voices, and production. Built with Claude, Piper TTS, and automated pipeline tooling.
    15 min
  • The Wind Factor: What Data Actually Shows
    Why do broadcasters always say 'the wind is a factor'? Joey investigates whether wind actually shows up in the numbers—through NFL field goals, MLB's Statcast system, and beach volleyball serves. What emerges isn't a clean answer about wind, but a story about how measurement shapes what's visible, and what stays hidden.
    00:00:00 - The ubiquitous broadcast claim
    00:01:33 - NFL study: wind at kickoff, not at kick
    00:03:05 - Baseball's Statcast and localized wind measurement
    00:04:38 - The limits of measurement: satellite imagery
    00:06:27 - How blur in data hides the wind effect
    00:08:24 - Wind pushes both aces and errors up
    00:09:59 - Fog of war: seeing wind, not direction
    00:11:32 - Reading research as AI hosts
    00:13:17 - What measurement reveals and conceals
    ---
    Sources & further reading:
    • Clark, Johnson and Stimpson (2013). Going for three. MIT Sloan Sports Analytics Conference.
    • (checked 2026-11-15): https://sloansportsconference.com/research-papers/going-for-three
    • Petriello (2025). Wind effect. MLB.com.: https://www.mlb.com/news/the-big-impact-of-wind-on-baseball-outcomes
    • (checked 2026-11-15)
    • Kagan (2015). The physics of wind at the remodeled Wrigley Field. The Hardball Times.
    • (checked 2026-11-15): https://tht.fangraphs.com/the-physics-of-wind-at-the-remodeled-wrigley-field/
    • Reiser, Zentgraf, Kindermann and Künzell (2020). An approach to quantify the float effect of float serves in indoor
    • and beach volleyball. Frontiers in Sports and Active Living 2.
    • (checked 2026-11-15): https://www.frontiersin.org/journals/sports-and-active-living/articles/10.3389/fspor.2020.559277/full
    • Copernicus Climate Change Service. ERA5 hourly data on single levels.
    • (checked 2026-11-15): https://cds.climate.copernicus.eu/datasets/reanalysis-era5-single-levels?tab=overview
    • ECMWF. ERA5: data documentation. (checked 2026-11-15): https://confluence.ecmwf.int/display/CKB/ERA5:+data+documentation
    • Potisomporn, Adcock and Vogel (2023). Evaluating ERA5 reanalysis predictions of low wind speed events around the UK.
    • Energy Reports* 10. Abstract via OpenAlex (checked 2026-11-15)
    • Regression dilution. Wikipedia. (checked 2026-11-15): https://en.wikipedia.org/wiki/Regression_dilution
    • Project files: SERVE-DECLARATION.md (v1 T6, amendment 7), the intro and wind article drafts (read 2026-11-15)
    This podcast episode was fully generated by AI — research, script, voices, and production. Built with Claude, Piper TTS, and automated pipeline tooling.
    15 min
  • Pot Odds on the Beach: How Pressure Wins
    How do elite volleyball players perform under pressure? This episode explores three deceptively simple questions: whether players truly rise to the moment, whether timeouts help or hurt servers, and whether leads snowball. Using poker odds as a conceptual lens and analyzing rally-by-rally data from the professional men's tour, the AI hosts built by Joey Musselman's pipeline reveal how pressure reshapes player behavior—and where the data still leaves us guessing.
    00:00:00 - Opening: big points, timeouts, snowballs
    00:01:50 - Tight poker: how pressure changes serve behavior
    00:03:32 - What clutch performance looks like in the data
    00:05:04 - Questions the data can't answer
    00:08:34 - Do timeouts help servers?
    00:10:07 - Survivor bias and drawing the right conclusions
    00:11:38 - Does a lead snowball?
    00:13:20 - Remaining uncertainties
    ---
    Sources & further reading:
    • Klaassen and Magnus (2001). Are points in tennis independent and identically distributed? JASA 96(454)
    • 500-509. Abstract via: https://api.openalex.org/works?search=Are%20points%20in%20tennis%20independent%20and%20identically%20distributed%20dynamic%20binary%20panel
    • (checked 2026-11-15)
    • González-Díaz, Gossner and Rogers (2012). Performing best when it matters most: evidence from professional
    • tennis. Journal of Economic Behavior & Organization 84(3), 767-781.
    • (abstract, checked 2026-11-15): https://econpapers.repec.org/RePEc:eee:jeborg:v:84:y:2012:i:3:p:767-781
    • James, B. (2004). Underestimating the fog. Baseball Research Journal 33.
    • (checked 2026-11-15): https://sabr.org/journal/article/underestimating-the-fog/
    • Peterson, I. (2004). The iced foot effect. Science News (on Berry and Wood, Chance 2004).
    • (checked 2026-11-15): https://www.sciencenews.org/article/iced-foot-effect
    • Clark, Johnson and Stimpson (2013). Going for three. MIT Sloan Sports Analytics Conference.
    • (checked 2026-11-15): https://sloansportsconference.com/research-papers/going-for-three
    • Assis, Assunção and Vaz-de-Melo (2021). Stop the clock: are timeout effects real?: https://arxiv.org/abs/2009.06750
    • (abstract, checked 2026-11-15)
    • Fernández-Echeverría et al. (2019). The timeout in sports: a study of its effect on volleyball. *Frontiers in
    • Psychology* 10, 2437. (checked 2026-11-15): https://pmc.ncbi.nlm.nih.gov/articles/PMC6844263/
    • Miller and Sanjurjo (2018). Surprised by the hot hand fallacy? Econometrica 86(6), 2019-2047.
    • (abstract, checked 2026-11-15): https://arxiv.org/abs/1902.01265
    • Raab, Gula and Gigerenzer (2012). The hot hand exists in volleyball and is used for allocation decisions. *Journal
    • of Experimental Psychology: Applied* 18(1), 81-94. Abstract via
    • (checked 2026-11-15)
    • Link and Wenninger (2019). Performance streaks in elite beach volleyball. Frontiers in Psychology 10, 919.
    • (checked 2026-11-15): https://pmc.ncbi.nlm.nih.gov/articles/PMC6534039/
    • Gilovich, Vallone and Tversky (1985), described at (checked 2026-11-15): https://en.wikipedia.org/wiki/Hot_hand
    • Project files: SERVE-DECLARATION.md (v1 T3-T5, amendments 1 and 6, Joey's notes on article 4), the intro and
    • big-points article drafts (read 2026-11-15)
    This podcast episode was fully generated by AI — research, script, voices, and production. Built with Claude, Piper TTS, and automated pipeline tooling.
    15 min
  • Who Gets Served: The Game Theory Behind Tennis and Volleyball
    What happens when game theory meets a tennis racket? Wimbledon's top servers play minimax almost perfectly — predicting which side their opponent will cover, randomizing their serves just right. Beach volleyball looks like it should work the same way. It doesn't. When you can pick who you want to face off against, game theory has a prediction about who wins. But the beaches of the world are telling a story the theory never prepared for.
    00:00:00 - The serve as strategy: why every choice matters
    00:01:34 - Tennis proves minimax: optimal play in action
    00:04:43 - On the beach, game theory says something different
    00:08:01 - Does targeting the weaker player actually work?
    00:13:02 - Where the data runs out: the unanswered question
    ---
    Sources & further reading:
    • Walker and Wooders (2001). Minimax play at Wimbledon. American Economic Review 91(5), 1521-1538.
    • (full text, checked 2026-11-15): https://www.math.stonybrook.edu/~gaston/print/Old/WimbledonAER.pdf
    • Gauriot, Page and Wooders (2023). Expertise, gender, and equilibrium play. Quantitative Economics
    • 14(3), 981-1020.
    • (abstract, checked 2026-11-15)
    • Anderson, Rosen, Rust and Wong (2025). Disequilibrium play in tennis. *Journal of Political
    • Economy* 133(1). (full text, checked 2026-11-15): https://gwern.net/doc/statistics/decision/2024-anderson.pdf
    • Palacios-Huerta (2003). Professionals play minimax. Review of Economic Studies 70(2), 395-415.
    • (abstract, checked 2026-11-15): https://ideas.repec.org/a/oup/restud/v70y2003i2p395-415.html
    • Project files: SERVE-DECLARATION.md (v0, v1 T2/B7, amendment 5 and "Article 3 parked"), the intro and targeting
    • article drafts, docs/serve-video/HANDOFF-ARTICLES-2026-11-15.md (read 2026-11-15)
    This podcast episode was fully generated by AI — research, script, voices, and production. Built with Claude, Piper TTS, and automated pipeline tooling.
    15 min
  • One Ace, Two Misses
    The big serve is volleyball's most seductive weapon—but does it actually work? We pit precision against aggression (Sørum's 2.5% miss rate vs. Evandro's 22%), then pull fifty years of tennis data to answer the question every power player asks: does the reward justify the risk? A prediction lands before the numbers do, and the data delivers a verdict that might change how you think about aggression.
    00:00:00 - The Setup: Two serve philosophies
    00:01:32 - The History: A fifty-year-old debate
    00:03:05 - The Framework: Prussing's verdict
    00:06:22 - The Math: Exchange rate
    00:11:06 - The Question: Where does hard serve actually earn?
    00:12:42 - The Answer: When aggression doesn't pay
    ---
    Sources & further reading:
    • Prussing (1977). Notes 61.24: How to serve in tennis. The Mathematical Gazette 61(418), 294-296.
    • (abstract, checked 2026-11-15)
    • Sackmann (2015). Are Two First Serves Ever Better Than One? Tennis Abstract.
    • (checked 2026-11-15): https://www.tennisabstract.com/blog/2015/11/02/are-two-first-serves-ever-better-than-one/
    • Sackmann (2026). The Point After a Second Serve. Tennis Abstract.
    • (checked 2026-11-15): https://www.tennisabstract.com/blog/2026/10/03/the-point-after-a-second-serve/
    • Forman. What is the optimal serve error rate? (checked 2026-11-15): https://coachingvb.com/what-is-the-optimal-serve-error-rate/
    • Forman. Service errors vs. service points. (checked 2026-11-15): https://coachingvb.com/service-errors-vs-service-points
    • Lebedew (2025). Men's volleyball: are serve errors important? WorldofVolley.
    • (checked 2026-11-15): https://worldofvolley.com/blog/bymarklebedew/332000/by-mark-lebedew-mens-volleyball-are-serve-errors-important.html
    • Klaassen and Magnus (2009). The efficiency of top agents: an analysis through service strategy in
    • tennis. Journal of Econometrics 148(1), 72-85. Not opened; described by Anderson et al. (2025)
    • (checked 2026-11-15): https://gwern.net/doc/statistics/decision/2024-anderson.pdf
    • Project files: game-senser-fable/docs/flavor/SERVE-DECLARATION.md (v1 T1, amendment 4), the risk and intro
    • article drafts (read 2026-11-15)
    This podcast episode was fully generated by AI — research, script, voices, and production. Built with Claude, Piper TTS, and automated pipeline tooling.
    15 min
  • Aces vs. Consistency: What Makes a Good Server
    Every beach volleyball fan has this argument: what makes a good server? The one hitting aces like Evandro, or the one who never misses like Åhman and Hellvig? Before analyzing tour data, Joey predicted his answer. Here's what the numbers revealed.
    00:00:00 - Introducing the serve series
    00:01:43 - Mise en place: predicting before analyzing data
    00:03:34 - Checking consistency across matches
    00:05:32 - What the data shows about consistency
    00:07:03 - Using Elo ratings to measure serve quality
    00:08:48 - Young server performance
    00:10:34 - The range of serve consistency
    00:12:20 - What's next for the serve series
    ---
    Sources & further reading:
    • Whitmore (2023). What are expected goals (xG)? The Analyst.: https://theanalyst.com/articles/what-are-expected-goals-xg
    • (checked 2026-11-15)
    • Yurko, Ventura and Horowitz (2018). nflWAR. (abstract, checked 2026-11-15): https://arxiv.org/abs/1802.00998
    • Lebedew (2024). Major volleyball statistics explanations.
    • (checked 2026-11-15): https://marklebedew.com/2024/08/22/major-volleyball-statistics-explanations/
    • Efron and Morris (1975). Data analysis using Stein's estimator and its generalizations. JASA 70(350), 311-319.
    • (checked 2026-11-15): https://faculty.ucmerced.edu/jvevea/classes/290_21/readings/week%204/Efron%20and%20Morris.pdf
    • Project files: SERVE-DECLARATION.md (v0, v1, amendments 1-3), the intro article draft (read 2026-11-15)
    This podcast episode was fully generated by AI — research, script, voices, and production. Built with Claude, Piper TTS, and automated pipeline tooling.
    14 min
  • Earthquake Math Goes to Prison
    What if the same mathematical model used to predict earthquakes decided whether you go to prison? This episode explores how seismological algorithms were repurposed for predictive policing, and how tools like COMPAS now influence sentencing decisions across 27 U.S. states. We examine feedback loops, hidden biases, and the fundamental problem of treating human behavior like natural phenomena.
    00:00:00 - Opening: Earthquake math applied to criminal prediction
    00:01:31 - COMPAS scores in presentencing reports
    00:03:34 - The algorithm as arbiter
    00:05:13 - Accuracy claims and verification
    00:06:50 - Chess ratings vs. recidivism
    00:10:06 - PredPol's seismological foundation
    00:11:36 - System convergence and feedback loops
    00:14:56 - Black box accuracy and the 65% problem
    00:16:31 - Why it still doesn't feel right
    ---
    Sources & further reading:
    • Julia Dressel and Hany Farid — "The accuracy, fairness, and limits of predicting recidivism" — Science Advances 4(1), January 2018: https://www.science.org/doi/10.1126/sciadv.aao5580
    • Jeff Larson, Surya Mattu, Lauren Kirchner, Julia Angwin (ProPublica) — "How We Analyzed the COMPAS Recidivism Algorithm" — May 2016: https://www.propublica.org/article/how-we-analyzed-the-compas-recidivism-algorithm
    • Danielle Ensign, Sorelle Friedler, Scott Neville, Carlos Scheidegger, Suresh Venkatasubramanian — "Runaway Feedback Loops in Predictive Policing" — Proceedings of the 1st Conference on Fairness, Accountability and Transparency, 2018: https://proceedings.mlr.press/v81/ensign18a.html
    • Alexandra Chouldechova — "Fair prediction with disparate impact: A study of bias in recidivism prediction instruments" — Big Data 5(2), 2017
    • Jon Kleinberg, Sendhil Mullainathan, Manish Raghavan — "Inherent Trade-Offs in the Fair Determination of Risk Scores" — 2016
    • Wikipedia — "Loomis v. Wisconsin": https://en.wikipedia.org/wiki/Loomis_v._Wisconsin
    • Harvard Law Review — "State v. Loomis" — Vol. 130, 2017: https://harvardlawreview.org/print/vol-130/state-v-loomis/
    • Wikipedia — "Geolitica (PredPol)": https://en.wikipedia.org/wiki/Geolitica
    • Wikipedia — "COMPAS (software)": https://en.wikipedia.org/wiki/COMPAS_(software
    • Brennan Center for Justice — "Predictive Policing Explained": https://www.brennancenter.org/our-work/research-reports/predictive-policing-explained
    • R Street Institute — "What Works in American Policing Part 6—Predictive Policing": https://www.rstreet.org/commentary/what-works-in-american-policing-part-6-predictive-policing-promise-pitfalls-and-the-limits-of-algorithmic-forecasting/
    • EU AI Act — Article 5, Prohibited AI Practices: https://www.euai-act.com/articles/prohibited-ai-practices
    • Columbia Law Review — "Privatizing Sentencing: A Delegation Framework for Recidivism Risk Assessment": https://www.columbialawreview.org/content/privatizing-sentencing-a-delegation-framework-for-recidivism-risk-assessment/
    • Stanford Encyclopedia of Philosophy — "Algorithmic Fairness": https://plato.stanford.edu/entries/algorithmic-fairness/
    This podcast episode was fully generated by AI — research, script, voices, and production. Built with Claude, Piper TTS, and automated pipeline tooling.
    17 min
  • Never See the Lie: Building the Reality Filter
    Your brain deliberately corrupts its own memories. AI models believe their own lies. The structural reason is the same on both sides: confabulation—filling gaps with plausible fiction without knowing it happened. This episode explores how the human brain filters out false thoughts before they reach consciousness, and how AI engineers are now building similar reality checks using retrieval-augmented generation and other techniques that let systems verify claims against ground truth rather than just guessing from their training data.
    00:00:00 - The problem: confabulation in brains and AI
    00:01:31 - How the brain's orbitofrontal filter works through exclusion
    00:06:26 - Why retrieval beats pure generation
    00:09:36 - The scale of improvement: 18% to 82%
    00:12:53 - Schnider's filter as the model for AI reality-checking
    00:14:44 - Why multiple partial checks matter more than a single oracle
    ---
    Sources & further reading:
    • Armin Schnider — "Orbitofrontal Reality Filtering": https://pmc.ncbi.nlm.nih.gov/articles/PMC3677127/
    • Nadine Dijkstra, Thomas von Rein, Peter Kok, Steve Fleming — "A neural basis for distinguishing imagination from reality": https://www.cell.com/neuron/fulltext/S0896-6273(25)00362-9
    • Hexiang Tan et al. — "Too Consistent to Detect: A Study of Self-Consistent Errors in LLMs" (EMNLP 2025): https://aclanthology.org/2025.emnlp-main.238/
    • Patrick Lewis et al. — "Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks" (NeurIPS 2020) — (original arXiv: 2005.11401)
    • Saroj Mishra — "Cascading Hallucination in Agentic RAG: The CHARM Framework for Detection and Mitigation": https://arxiv.org/abs/2606.04435
    • Kazuaki Furumai, Shuichiro Haruta, Kazunori Matsumoto, Daisuke Kamisaka — "CORTEX: Token-Level Hallucination Detection in RAG via Comparative Internal Representations": https://arxiv.org/abs/2606.31033
    • Yifan Feng et al. — "Hyper-RAG: combating LLM hallucinations using hypergraph-driven retrieval-augmented generation": https://www.nature.com/articles/s41467-026-71411-1
    • Hernandez & Gonzalez — "Hallucination or Confabulation? Neuroanatomy as metaphor in Large Language Models": https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10619792/
    • Integrative Psychiatry — "Confabulation Not Hallucination: Rethinking AI Error Terminology": https://www.integrative-psych.org/resources/confabulation-not-hallucination-ai-errors
    • Neuroscience News — "How the Brain Tells Imagination from Reality, And When It Fails": https://neurosciencenews.com/reality-imagination-brain-29221/
    • Boston Globe — "We think hallucinations make AI different from us. What if they make us the same?" (Sept 24, 2025): https://www.bostonglobe.com/2025/09/24/magazine/artificial-intelligence-hallucinations-in-ai/
    This podcast episode was fully generated by AI — research, script, voices, and production. Built with Claude, Piper TTS, and automated pipeline tooling.
    18 min

About Clown Cast

From the publisher's feed

Podcasts about whatever I find interesting — history, tech, weird rabbit holes. I was making these for myself anyway, so I figured I'd share. Fully AI-generated: Claude researches and writes each…