{"id":"02-two-systems","number":2,"slug":"system-one-and-system-two","title":"System One and System Two","short":"System One and System Two","module":"The idea","summary":"The name comes from Daniel Kahneman: fast, automatic judgment versus slow, effortful reasoning. Here is what the metaphor gets right about decision models, where it breaks, and the rule of thumb it gives you: if an expert could answer in a few seconds, it is a System One question.","voice":"Jessica","scenes":[{"template":"title","chapter":"Intro","narration":"Why call a model System One? The name comes from psychology, and it hands you the most useful rule in this course: which questions belong to a fast model, and which don't.","params":{"kicker":"Lesson 2","module":"Module 1 · The idea","title":"System One and *System Two*","subtitle":"Where the name comes from, and the rule of thumb it gives you.","cast":["snap",{"who":"mull","mood":"think"}]},"start":0.5,"dur":12.276,"words":[{"text":"Why","s":0.8,"e":1.04},{"text":"call","s":1.12,"e":1.28},{"text":"a","s":1.36,"e":1.44},{"text":"model","s":1.52,"e":1.84},{"text":"System","s":2,"e":2.4},{"text":"One?","s":2.48,"e":2.88},{"text":"The","s":3.36,"e":3.4},{"text":"name","s":3.44,"e":3.68},{"text":"comes","s":3.76,"e":4},{"text":"from","s":4.08,"e":4.16},{"text":"psychology,","s":4.32,"e":5.2},{"text":"and","s":5.24,"e":5.28},{"text":"it","s":5.36,"e":5.44},{"text":"hands","s":5.52,"e":5.76},{"text":"you","s":5.8,"e":5.84},{"text":"the","s":5.92,"e":6},{"text":"most","s":6.16,"e":6.48},{"text":"useful","s":6.56,"e":6.88},{"text":"rule","s":6.96,"e":7.12},{"text":"in","s":7.2,"e":7.28},{"text":"this","s":7.36,"e":7.44},{"text":"course:","s":7.52,"e":8.48},{"text":"which","s":8.6,"e":8.72},{"text":"questions","s":8.8,"e":9.2},{"text":"belong","s":9.28,"e":9.52},{"text":"to","s":9.6,"e":9.68},{"text":"a","s":9.76,"e":9.8},{"text":"fast","s":9.84,"e":10.16},{"text":"model,","s":10.24,"e":10.8},{"text":"and","s":10.88,"e":10.96},{"text":"which","s":11.08,"e":11.2},{"text":"don't.","s":11.28,"e":11.92}]},{"template":"compare","chapter":"Two ways to think","narration":"Try this. What's two plus two? The answer just appears. Now, what's seventeen times twenty four? You can feel yourself start to work. Daniel Kahneman, who won a Nobel prize for his work on judgment and decision making, called these System 1 and System 2.","params":{"left":{"title":"2 + 2 = ?","who":"snap","tone":"good","items":[{"text":"The answer just appears","at":"just appears"},{"text":"Automatic, effortless"}],"at":"two plus two"},"right":{"title":"17 × 24 = ?","who":"mull","tone":"s2","items":[{"text":"You feel yourself start to work","at":"start to work"},{"text":"Serial, effortful"}],"at":"seventeen"},"vs":"vs"},"start":12.776,"dur":19.079,"words":[{"text":"Try","s":13.076,"e":13.316},{"text":"this.","s":13.396,"e":13.916},{"text":"What's","s":14.116,"e":14.516},{"text":"two","s":14.596,"e":14.756},{"text":"plus","s":14.836,"e":15.076},{"text":"two?","s":15.236,"e":15.676},{"text":"The","s":15.956,"e":16.036},{"text":"answer","s":16.116,"e":16.436},{"text":"just","s":16.676,"e":16.836},{"text":"appears.","s":16.916,"e":17.756},{"text":"Now,","s":18.036,"e":18.476},{"text":"what's","s":18.676,"e":18.916},{"text":"seventeen","s":18.996,"e":19.556},{"text":"times","s":19.636,"e":19.956},{"text":"twenty","s":20.116,"e":20.276},{"text":"four?","s":20.436,"e":20.996},{"text":"You","s":21.156,"e":21.236},{"text":"can","s":21.396,"e":21.556},{"text":"feel","s":21.636,"e":21.876},{"text":"yourself","s":21.956,"e":22.356},{"text":"start","s":22.476,"e":22.596},{"text":"to","s":22.676,"e":22.756},{"text":"work.","s":22.956,"e":23.436},{"text":"Daniel","s":23.716,"e":24.036},{"text":"Kahneman,","s":24.116,"e":24.716},{"text":"who","s":24.756,"e":24.916},{"text":"won","s":24.996,"e":25.076},{"text":"a","s":25.156,"e":25.236},{"text":"Nobel","s":25.276,"e":25.556},{"text":"prize","s":25.636,"e":26.036},{"text":"for","s":26.076,"e":26.116},{"text":"his","s":26.196,"e":26.356},{"text":"work","s":26.436,"e":26.516},{"text":"on","s":26.596,"e":26.756},{"text":"judgment","s":26.836,"e":27.236},{"text":"and","s":27.276,"e":27.316},{"text":"decision","s":27.396,"e":27.716},{"text":"making,","s":27.796,"e":28.476},{"text":"called","s":28.516,"e":28.756},{"text":"these","s":28.836,"e":29.156},{"text":"System","s":29.236,"e":29.636},{"text":"1","s":29.716,"e":30.036},{"text":"and","s":30.116,"e":30.196},{"text":"System","s":30.276,"e":30.596},{"text":"2.","s":30.756,"e":31.156}]},{"template":"quote","chapter":"Two ways to think","narration":"In his Nobel lecture he described them like this. System 1 is fast, automatic, effortless and associative. System 2 is slower, serial, effortful and deliberately controlled.","params":{"text":"The operations of System 1 are *fast, automatic, effortless, associative*… The operations of System 2 are slower, serial, effortful, and deliberately controlled.","who":"Daniel Kahneman","role":"Nobel lecture, 2002","src":"Maps of Bounded Rationality"},"start":31.855,"dur":14.429,"words":[{"text":"In","s":32.155,"e":32.315},{"text":"his","s":32.395,"e":32.475},{"text":"Nobel","s":32.555,"e":32.875},{"text":"lecture","s":32.955,"e":33.355},{"text":"he","s":33.515,"e":33.675},{"text":"described","s":33.755,"e":34.075},{"text":"them","s":34.155,"e":34.235},{"text":"like","s":34.315,"e":34.475},{"text":"this.","s":34.555,"e":35.235},{"text":"System","s":35.515,"e":35.835},{"text":"1","s":35.995,"e":36.235},{"text":"is","s":36.315,"e":36.475},{"text":"fast,","s":36.635,"e":37.275},{"text":"automatic,","s":37.595,"e":38.155},{"text":"effortless","s":38.235,"e":38.795},{"text":"and","s":38.875,"e":38.955},{"text":"associative.","s":39.035,"e":40.075},{"text":"System","s":40.315,"e":40.635},{"text":"2","s":40.795,"e":41.115},{"text":"is","s":41.275,"e":41.435},{"text":"slower,","s":41.515,"e":42.315},{"text":"serial,","s":42.395,"e":43.115},{"text":"effortful","s":43.195,"e":43.755},{"text":"and","s":43.915,"e":43.995},{"text":"deliberately","s":44.075,"e":44.715},{"text":"controlled.","s":44.795,"e":45.675}]},{"template":"quote","chapter":"Fast is not dumb","narration":"And System 1 isn't the dumb one. Kahneman's point is that this is where skill lives. A chess master's strong moves come to mind instantly. System 2 just checks them.","params":{"text":"System 1 is *where skill is*. That is when we get to be skilled at something it becomes automatic.","who":"Daniel Kahneman","role":"Talks at Google","src":"Thinking, Fast and Slow","t":1000},"start":46.284,"dur":12.183,"words":[{"text":"And","s":46.584,"e":46.717},{"text":"System","s":46.744,"e":47.157},{"text":"1","s":47.224,"e":47.464},{"text":"isn't","s":47.544,"e":48.024},{"text":"the","s":48.104,"e":48.184},{"text":"dumb","s":48.264,"e":48.504},{"text":"one.","s":48.584,"e":48.904},{"text":"Kahneman's","s":49.304,"e":49.784},{"text":"point","s":49.944,"e":50.104},{"text":"is","s":50.184,"e":50.264},{"text":"that","s":50.344,"e":50.504},{"text":"this","s":50.584,"e":50.744},{"text":"is","s":50.824,"e":50.904},{"text":"where","s":50.984,"e":51.064},{"text":"skill","s":51.224,"e":51.624},{"text":"lives.","s":51.704,"e":52.304},{"text":"A","s":52.504,"e":52.584},{"text":"chess","s":52.664,"e":52.824},{"text":"master's","s":52.904,"e":53.384},{"text":"strong","s":53.584,"e":53.784},{"text":"moves","s":53.864,"e":54.184},{"text":"come","s":54.264,"e":54.344},{"text":"to","s":54.424,"e":54.504},{"text":"mind","s":54.624,"e":54.744},{"text":"instantly.","s":54.904,"e":55.784},{"text":"System","s":56.024,"e":56.264},{"text":"2","s":56.424,"e":56.664},{"text":"just","s":56.824,"e":56.984},{"text":"checks","s":57.064,"e":57.304},{"text":"them.","s":57.384,"e":57.864}]},{"template":"spectrum","chapter":"Models on the spectrum","narration":"Models sit on the same spectrum. A decision model reads the state once and answers in a single pass, in a fraction of a second. Reasoning models do the opposite: they spend tokens thinking, step by step, and trade time for accuracy.","params":{"title":"From one pass to many steps","left":{"label":"System One","sub":"one pass, ~0.25 s","who":"snap"},"right":{"label":"System Two","sub":"thinks step by step, seconds to minutes","who":"mull"},"marks":[{"pos":0.08,"label":"Is this a refund request?","at":"single pass"},{"pos":0.3,"label":"Which team owns it?"},{"pos":0.62,"label":"Draft the reply","at":"Reasoning models"},{"pos":0.92,"label":"Plan the migration","at":"trade time"}]},"start":58.467,"dur":16.519,"words":[{"text":"Models","s":58.767,"e":59.28},{"text":"sit","s":59.327,"e":59.514},{"text":"on","s":59.567,"e":59.627},{"text":"the","s":59.647,"e":59.78},{"text":"same","s":59.807,"e":60.127},{"text":"spectrum.","s":60.207,"e":60.847},{"text":"A","s":61.247,"e":61.327},{"text":"decision","s":61.567,"e":61.807},{"text":"model","s":61.887,"e":62.127},{"text":"reads","s":62.207,"e":62.447},{"text":"the","s":62.527,"e":62.607},{"text":"state","s":62.767,"e":62.927},{"text":"once","s":63.007,"e":63.407},{"text":"and","s":63.487,"e":63.647},{"text":"answers","s":63.847,"e":64.047},{"text":"in","s":64.127,"e":64.167},{"text":"a","s":64.207,"e":64.367},{"text":"single","s":64.527,"e":64.687},{"text":"pass,","s":64.767,"e":65.327},{"text":"in","s":65.367,"e":65.407},{"text":"a","s":65.487,"e":65.647},{"text":"fraction","s":65.847,"e":66.047},{"text":"of","s":66.127,"e":66.167},{"text":"a","s":66.207,"e":66.247},{"text":"second.","s":66.287,"e":67.247},{"text":"Reasoning","s":67.447,"e":67.727},{"text":"models","s":67.927,"e":68.127},{"text":"do","s":68.207,"e":68.287},{"text":"the","s":68.327,"e":68.367},{"text":"opposite:","s":68.447,"e":69.247},{"text":"they","s":69.327,"e":69.487},{"text":"spend","s":69.527,"e":69.807},{"text":"tokens","s":70.047,"e":70.287},{"text":"thinking,","s":70.367,"e":71.087},{"text":"step","s":71.167,"e":71.407},{"text":"by","s":71.487,"e":71.647},{"text":"step,","s":71.727,"e":72.207},{"text":"and","s":72.287,"e":72.367},{"text":"trade","s":72.527,"e":72.687},{"text":"time","s":72.927,"e":73.167},{"text":"for","s":73.247,"e":73.327},{"text":"accuracy.","s":73.407,"e":74.287}]},{"template":"statement","chapter":"The rule of thumb","narration":"So here's the rule of thumb, straight from the TypeSafe docs. Make each question a gut check: the kind of judgment a knowledgeable person could make in a few seconds, given the right context.","params":{"eyebrow":"The rule of thumb","text":"If an expert could answer it *in a few seconds*, it's a System One question.","art":"snap","mood":"sure","punch":"in a few seconds"},"start":74.986,"dur":12.339,"words":[{"text":"So","s":75.286,"e":75.526},{"text":"here's","s":75.686,"e":75.926},{"text":"the","s":75.966,"e":76.006},{"text":"rule","s":76.126,"e":76.246},{"text":"of","s":76.286,"e":76.326},{"text":"thumb,","s":76.406,"e":76.966},{"text":"straight","s":77.046,"e":77.366},{"text":"from","s":77.446,"e":77.526},{"text":"the","s":77.566,"e":77.606},{"text":"TypeSafe","s":77.686,"e":78.166},{"text":"docs.","s":78.406,"e":78.926},{"text":"Make","s":79.206,"e":79.446},{"text":"each","s":79.526,"e":79.686},{"text":"question","s":79.846,"e":80.326},{"text":"a","s":80.406,"e":80.486},{"text":"gut","s":80.646,"e":80.726},{"text":"check:","s":80.886,"e":81.526},{"text":"the","s":81.566,"e":81.606},{"text":"kind","s":81.686,"e":81.846},{"text":"of","s":81.886,"e":81.926},{"text":"judgment","s":82.006,"e":82.406},{"text":"a","s":82.486,"e":82.566},{"text":"knowledgeable","s":82.646,"e":83.126},{"text":"person","s":83.206,"e":83.526},{"text":"could","s":83.606,"e":83.766},{"text":"make","s":83.886,"e":84.006},{"text":"in","s":84.046,"e":84.086},{"text":"a","s":84.166,"e":84.246},{"text":"few","s":84.326,"e":84.486},{"text":"seconds,","s":84.566,"e":85.286},{"text":"given","s":85.406,"e":85.526},{"text":"the","s":85.566,"e":85.606},{"text":"right","s":85.686,"e":85.766},{"text":"context.","s":85.846,"e":86.726}]},{"template":"compare","chapter":"The rule of thumb","narration":"Is this customer asking for a refund? Which team should handle it? How severe is this bug? An expert answers those at a glance. But should we make an exception for this customer, given three years of history? Write the apology? Plan the rollout? Those need System Two.","params":{"left":{"title":"System One questions","who":"snap","tone":"good","items":[{"text":"Is the customer asking for a refund?","at":"asking for a refund"},{"text":"Which team should handle it?","at":"Which team"},{"text":"How severe is this bug?","at":"How severe"}]},"right":{"title":"System Two work","who":"mull","tone":"s2","items":[{"text":"Make an exception, given 3 years of history?","at":"exception"},{"text":"Write the apology","at":"apology"},{"text":"Plan the rollout","at":"rollout"}],"at":"But should we"},"vs":"vs"},"start":87.325,"dur":18.191,"words":[{"text":"Is","s":87.625,"e":87.745},{"text":"this","s":87.785,"e":87.985},{"text":"customer","s":88.025,"e":88.455},{"text":"asking","s":88.505,"e":88.852},{"text":"for","s":88.905,"e":89.038},{"text":"a","s":89.065,"e":89.105},{"text":"refund?","s":89.145,"e":90.105},{"text":"Which","s":90.185,"e":90.425},{"text":"team","s":90.545,"e":90.665},{"text":"should","s":90.745,"e":90.825},{"text":"handle","s":90.905,"e":91.145},{"text":"it?","s":91.225,"e":91.865},{"text":"How","s":91.945,"e":92.025},{"text":"severe","s":92.105,"e":92.585},{"text":"is","s":92.665,"e":92.745},{"text":"this","s":92.825,"e":92.905},{"text":"bug?","s":92.985,"e":93.785},{"text":"An","s":93.865,"e":93.945},{"text":"expert","s":94.145,"e":94.345},{"text":"answers","s":94.425,"e":94.825},{"text":"those","s":94.985,"e":95.145},{"text":"at","s":95.225,"e":95.305},{"text":"a","s":95.385,"e":95.425},{"text":"glance.","s":95.465,"e":96.345},{"text":"But","s":96.465,"e":96.585},{"text":"should","s":96.745,"e":96.905},{"text":"we","s":96.985,"e":97.065},{"text":"make","s":97.145,"e":97.225},{"text":"an","s":97.265,"e":97.305},{"text":"exception","s":97.585,"e":97.865},{"text":"for","s":97.945,"e":98.025},{"text":"this","s":98.105,"e":98.185},{"text":"customer,","s":98.265,"e":98.745},{"text":"given","s":98.825,"e":99.065},{"text":"three","s":99.145,"e":99.385},{"text":"years","s":99.465,"e":99.625},{"text":"of","s":99.705,"e":99.785},{"text":"history?","s":100.025,"e":100.745},{"text":"Write","s":100.825,"e":100.905},{"text":"the","s":100.985,"e":101.038},{"text":"apology?","s":101.092,"e":101.945},{"text":"Plan","s":102.065,"e":102.185},{"text":"the","s":102.225,"e":102.265},{"text":"rollout?","s":102.385,"e":103.145},{"text":"Those","s":103.305,"e":103.465},{"text":"need","s":103.545,"e":103.705},{"text":"System","s":103.865,"e":104.185},{"text":"Two.","s":104.345,"e":104.905}]},{"template":"flow","chapter":"The layer cake","narration":"Developer Daniel Sam summed up the architecture in one line: put a decision layer between your rules and your reasoner. Let code be exact, let the model judge, let the LLM think.","params":{"title":"Let code be exact, let the model judge, let the LLM think","nodes":[{"label":"Code","sub":"exact: rules, math, dates","kind":"curly","at":"code be exact","edge":"narrow questions"},{"label":"Decision model","sub":"judges: typed answers","kind":"snap","at":"model judge","edge":"hard cases"},{"label":"LLM or person","sub":"thinks: reasoning, writing","kind":"mull","at":"LLM think"}]},"start":105.516,"dur":13.776,"words":[{"text":"Developer","s":105.816,"e":106.296},{"text":"Daniel","s":106.456,"e":106.776},{"text":"Sam","s":106.856,"e":107.256},{"text":"summed","s":107.336,"e":107.576},{"text":"up","s":107.656,"e":107.736},{"text":"the","s":107.776,"e":107.816},{"text":"architecture","s":107.896,"e":108.536},{"text":"in","s":108.616,"e":108.776},{"text":"one","s":108.856,"e":109.016},{"text":"line:","s":109.096,"e":109.976},{"text":"put","s":110.056,"e":110.136},{"text":"a","s":110.176,"e":110.216},{"text":"decision","s":110.296,"e":110.696},{"text":"layer","s":110.776,"e":111.016},{"text":"between","s":111.096,"e":111.416},{"text":"your","s":111.496,"e":111.576},{"text":"rules","s":111.656,"e":112.136},{"text":"and","s":112.216,"e":112.376},{"text":"your","s":112.416,"e":112.456},{"text":"reasoner.","s":112.536,"e":113.336},{"text":"Let","s":113.576,"e":113.736},{"text":"code","s":113.816,"e":114.136},{"text":"be","s":114.216,"e":114.296},{"text":"exact,","s":114.576,"e":115.176},{"text":"let","s":115.256,"e":115.336},{"text":"the","s":115.376,"e":115.416},{"text":"model","s":115.576,"e":115.896},{"text":"judge,","s":116.056,"e":116.856},{"text":"let","s":116.936,"e":117.016},{"text":"the","s":117.096,"e":117.176},{"text":"LLM","s":117.336,"e":117.816},{"text":"think.","s":118.056,"e":118.696}]},{"template":"statement","chapter":"Where the metaphor breaks","narration":"One caution. Kahneman himself called the two systems fictitious characters, a way to talk, not two boxes in the brain. And System 1 thinking is famous for its biases. A fast model can be confidently wrong too. What makes fast judgments usable in software is honest uncertainty, which is module three of this course.","params":{"eyebrow":"Kahneman, Talks at Google","text":"“Those are *fictitious characters*. They don't exist.”","sub":"Fast can still be wrong. What makes it usable is honest uncertainty.","sub_at":"honest uncertainty","art":"snap","mood":"unsure","react":{"at":"confidently wrong","mood":"unsure"}},"start":119.292,"dur":23.076,"words":[{"text":"One","s":119.592,"e":119.792},{"text":"caution.","s":119.832,"e":120.472},{"text":"Kahneman","s":120.552,"e":121.352},{"text":"himself","s":121.512,"e":121.992},{"text":"called","s":122.112,"e":122.232},{"text":"the","s":122.272,"e":122.312},{"text":"two","s":122.392,"e":122.552},{"text":"systems","s":122.632,"e":123.112},{"text":"fictitious","s":123.272,"e":123.912},{"text":"characters,","s":123.992,"e":125.032},{"text":"a","s":125.072,"e":125.112},{"text":"way","s":125.192,"e":125.272},{"text":"to","s":125.352,"e":125.432},{"text":"talk,","s":125.512,"e":126.232},{"text":"not","s":126.312,"e":126.552},{"text":"two","s":126.712,"e":126.872},{"text":"boxes","s":126.952,"e":127.352},{"text":"in","s":127.392,"e":127.432},{"text":"the","s":127.512,"e":127.552},{"text":"brain.","s":127.592,"e":127.992},{"text":"And","s":128.472,"e":128.552},{"text":"System","s":128.632,"e":128.952},{"text":"1","s":129.032,"e":129.192},{"text":"thinking","s":129.272,"e":129.672},{"text":"is","s":129.752,"e":129.912},{"text":"famous","s":130.152,"e":130.552},{"text":"for","s":130.632,"e":130.712},{"text":"its","s":130.792,"e":130.872},{"text":"biases.","s":130.952,"e":131.592},{"text":"A","s":131.992,"e":132.072},{"text":"fast","s":132.152,"e":132.472},{"text":"model","s":132.552,"e":132.792},{"text":"can","s":132.872,"e":132.952},{"text":"be","s":133.032,"e":133.112},{"text":"confidently","s":133.192,"e":133.832},{"text":"wrong","s":133.912,"e":134.312},{"text":"too.","s":134.392,"e":135.072},{"text":"What","s":135.352,"e":135.432},{"text":"makes","s":135.512,"e":135.752},{"text":"fast","s":135.832,"e":136.072},{"text":"judgments","s":136.152,"e":136.552},{"text":"usable","s":136.632,"e":136.952},{"text":"in","s":137.032,"e":137.112},{"text":"software","s":137.192,"e":137.752},{"text":"is","s":137.832,"e":138.072},{"text":"honest","s":138.152,"e":138.552},{"text":"uncertainty,","s":138.632,"e":139.672},{"text":"which","s":139.752,"e":139.832},{"text":"is","s":139.872,"e":139.912},{"text":"module","s":139.992,"e":140.312},{"text":"three","s":140.472,"e":140.632},{"text":"of","s":140.672,"e":140.712},{"text":"this","s":140.792,"e":140.952},{"text":"course.","s":141.032,"e":141.752}]},{"template":"quiz","chapter":"Quick check","narration":"Quick check. Which of these is a System One question?","params":{"q":"Which of these is a **System One** question?","options":["Given this contract and our policy history, should we waive the penalty?","Does this message mention a delivery problem?","Write a migration plan for this database","How many invoices in this list are overdue?"],"answer":1,"explain":"An expert spots a delivery problem at a glance. Waiving a penalty and planning a migration need reasoning or writing, and counting overdue invoices is exact work for code."},"start":142.368,"dur":9.053,"words":[{"text":"Quick","s":142.668,"e":142.908},{"text":"check.","s":142.988,"e":143.508},{"text":"Which","s":143.708,"e":143.868},{"text":"of","s":143.948,"e":144.028},{"text":"these","s":144.108,"e":144.348},{"text":"is","s":144.428,"e":144.508},{"text":"a","s":144.588,"e":144.668},{"text":"System","s":144.868,"e":145.068},{"text":"One","s":145.148,"e":145.308},{"text":"question?","s":145.468,"e":146.268}]},{"template":"exercise","chapter":"Try it","narration":"Try it. Take the list of hidden decisions from the last lesson and have your agent sort them into System One questions and System Two work.","params":{"title":"Sort your decisions","intro":"Paste this into your coding agent, with last lesson's list:","prompt":"For each LLM decision we found in this codebase, classify it:\n- System One: an expert could answer it in a few seconds from the given data\n- System Two: it needs multi-step reasoning, new text, or planning\n- Code: it is exact (math, dates, counting, lookups)\nFor every System Two item, propose how to split it into narrow\nquestions plus code, or say why it should stay with an LLM."},"start":151.421,"dur":9.231,"words":[{"text":"Try","s":151.721,"e":151.988},{"text":"it.","s":152.041,"e":152.281},{"text":"Take","s":152.681,"e":152.921},{"text":"the","s":152.961,"e":153.001},{"text":"list","s":153.081,"e":153.241},{"text":"of","s":153.321,"e":153.401},{"text":"hidden","s":153.481,"e":153.721},{"text":"decisions","s":153.841,"e":154.281},{"text":"from","s":154.361,"e":154.441},{"text":"the","s":154.481,"e":154.521},{"text":"last","s":154.601,"e":154.841},{"text":"lesson","s":154.921,"e":155.321},{"text":"and","s":155.481,"e":155.721},{"text":"have","s":155.801,"e":155.881},{"text":"your","s":155.961,"e":156.041},{"text":"agent","s":156.121,"e":156.441},{"text":"sort","s":156.521,"e":156.761},{"text":"them","s":156.881,"e":157.001},{"text":"into","s":157.121,"e":157.241},{"text":"System","s":157.321,"e":157.641},{"text":"One","s":157.721,"e":157.881},{"text":"questions","s":157.961,"e":158.521},{"text":"and","s":158.601,"e":158.761},{"text":"System","s":158.841,"e":159.161},{"text":"Two","s":159.241,"e":159.401},{"text":"work.","s":159.481,"e":160.041}]},{"template":"recap","chapter":"Recap","narration":"So: System One is fast, automatic judgment, and that's what decision models aim for. If an expert could answer in seconds, it's a System One question. Code stays exact, and System Two takes the hard cases. Next: what's actually inside a decision model.","params":{"items":["System 1: fast, automatic. System 2: slow, effortful.","An expert could answer in seconds? It's a System One question.","Code is exact, the model judges, the LLM thinks.","Fast can be wrong: honest uncertainty is what makes it usable."],"next":"How a decision model works","next_at":"Next"},"start":160.652,"dur":20.029,"words":[{"text":"So:","s":160.952,"e":161.352},{"text":"System","s":161.752,"e":162.099},{"text":"One","s":162.152,"e":162.472},{"text":"is","s":162.552,"e":162.712},{"text":"fast,","s":162.792,"e":163.512},{"text":"automatic","s":163.792,"e":164.072},{"text":"judgment,","s":164.152,"e":164.872},{"text":"and","s":164.952,"e":165.032},{"text":"that's","s":165.112,"e":165.325},{"text":"what","s":165.379,"e":165.432},{"text":"decision","s":165.632,"e":165.832},{"text":"models","s":165.912,"e":166.232},{"text":"aim","s":166.352,"e":166.472},{"text":"for.","s":166.552,"e":167.272},{"text":"If","s":167.352,"e":167.432},{"text":"an","s":167.472,"e":167.512},{"text":"expert","s":167.592,"e":167.912},{"text":"could","s":167.992,"e":168.072},{"text":"answer","s":168.232,"e":168.392},{"text":"in","s":168.472,"e":168.552},{"text":"seconds,","s":168.632,"e":169.352},{"text":"it's","s":169.392,"e":169.512},{"text":"a","s":169.552,"e":169.592},{"text":"System","s":169.752,"e":169.912},{"text":"One","s":169.992,"e":170.152},{"text":"question.","s":170.232,"e":171.192},{"text":"Code","s":171.272,"e":171.592},{"text":"stays","s":171.672,"e":171.912},{"text":"exact,","s":171.992,"e":172.792},{"text":"and","s":172.872,"e":172.952},{"text":"System","s":173.032,"e":173.272},{"text":"Two","s":173.432,"e":173.672},{"text":"takes","s":173.752,"e":173.992},{"text":"the","s":174.032,"e":174.072},{"text":"hard","s":174.152,"e":174.312},{"text":"cases.","s":174.392,"e":175.512},{"text":"Next:","s":175.592,"e":176.312},{"text":"what's","s":176.392,"e":176.712},{"text":"actually","s":176.952,"e":177.192},{"text":"inside","s":177.272,"e":177.832},{"text":"a","s":177.912,"e":177.952},{"text":"decision","s":177.992,"e":178.392},{"text":"model.","s":178.592,"e":179.032}]}],"total":181.081,"audio":"/audio/02-two-systems.mp3"}