Would you kill baby Hitler if you could go back in time? Is evil really a birth trait or shaped by nurture?

Shaw argues against labeling people as inherently evil. She suggests traits are shaped by nurture and environment rather than being an intrinsic, unchanging essence. The baby-Hitler discussion centers on whether birth or upbringing explains extreme harm (timestamp ~228).

Why should we empathize with people labeled as evil or monstrous?

Empathy helps us understand the pathways leading to harmful behavior and improve safety by addressing underlying causes rather than simply condemning individuals. The case for empathy is presented during the talk about empathy toward those labeled as evil (timestamp ~320).

Can you trust lie detection in policing? Are officers good at detecting deception?

Evidence shows deception detection is hard; officers are often overconfident in their ability to tell when someone is lying, and experiments show they perform near chance. Shaw discusses this limitation and its implications for policing (timestamp ~903).

What is the ‘heroic imagination’ and why is it important?

The heroic imagination prompts people to picture stepping up and interrupting danger in public, helping prevent bystander paralysis. Shaw discusses this concept in the context of real-world heroism and the Stanford Prison Experiment (timestamp ~2000).

What is false memory, and how easily can memories be implanted?

False memories are common and can be implanted with leading questions and imagination techniques. Shaw details experiments showing high susceptibility, including a 70% figure in a memory-implant study (timestamp ~1170).

How can AI affect memories and memory research?

AI can distort memory by generating persuasive but false content, influencing how people remember events. Shaw discusses AI as a potential false-memory machine and the need for safeguards (timestamp ~9719).

What is the cognitive interview and how is it used with AI?

The cognitive interview is a scripted memory interview protocol; Shaw describes adapting it for AI to elicit accurate memories without leading witnesses (timestamp ~9429).

What is Spot and how can it help preserve memories or evidence?

Spot is a platform to record important emotional memories and share them, originally stemming from Shaw's cognitive-interview work and expanding into workplace compliance. The founder discussion begins around the Spot description (timestamp ~9365).

What is the Tinder Swindler case about and what does it teach about fraud?

The Tinder Swindler case demonstrates how love bombing and tailored deception can defraud victims; Shaw highlights how scammers exploit the desire to be loved, and how victims rationalize or overlook red flags (timestamp ~2719).

What is Green Crime and why frame environmental harm as a ‘crime against Earth’?

Green Crime applies criminal-psychology insights to environmental harms (e.g., poaching, illegal mining, corporate fraud) to understand perpetrators and improve enforcement; Shaw frames it as a societal crime against Earth (timestamp ~8313).

Julia Shaw: Criminal Psychology of Murder, Serial Killers, Memory & Sex | Lex Fridman Podcast #483

Lex Fridman

Science & Technology4 min read163 min video

Oct 14, 2025|381,191 views|6,216|1,639

Julia Shaw alex friedman lex ai lex debate lex freedman lex fridman lex friedman lex interview lex lecture lex mit lex podcast lex transcript

Save to Pod

Key Moments

TL;DR

Evil is a spectrum; empathy, memory, and environment shape crime; beware bias and manipulation.

Key Insights

Dark traits are a spectrum: psychopathy, Machiavellianism, narcissism, and sadism form a continuum, not a binary ‘monster or not’ label; people can have subclinical levels and still influence risk.

Empathy as a compass: understanding the root causes of harmful behavior through empathy helps prevent harm and avoids dehumanizing those labeled as 'evil'.

Memory is fallible: deception detection is hard, eyewitness memory can be distorted, and expert input on memory can prevent wrongful conclusions in trials.

Murder motives are often ordinary and impulsive: many murders arise from small triggers or fights, not grandiose evil plans; recidivism for homicide is relatively low compared to other crimes.

Bystander effect and heroic imagination: simulation and practice of stepping in can increase real-world intervention; the ‘heroic imagination’ helps counteract passivity in crises.

Modern risks: online echo chambers fuel entitlement and violence (incel culture); frauds like the Tinder Swindler show how emotional manipulation can be weaponized, with AI amplifying capabilities.

THE DARK TETRAD: A CONTINUUM OF EVIL

Julia Shaw explains that the so-called dark tetrad—psychopathy, Machiavellianism, narcissism, and sadism—exists on a spectrum rather than as a binary label. Each trait has a scale, and people can be low on some and high on others, often remaining subclinical. When someone scores high across multiple traits, they’re more likely to engage in risky or harmful behaviors. This reframes evil as a human capacity that exists in varying degrees across all people, challenging the intuition of monsters and prompting more nuanced risk assessment and prevention.

NATURE, NURTURE, AND THE HITLER QUESTION

The conversation turns to the classic thought experiment: would you kill baby Hitler? Shaw argues this probes whether people are born evil or shaped by environment. She leans toward nurture, noting Hitler’s early life didn’t reveal the monstrous traits he later exhibited. The key takeaway is to resist labeling people as inherently evil; recognizing the mutable nature of behavior helps us understand factors that lead to atrocity and drives us toward prevention rather than dehumanization.

EMPATHY AS A TOOL FOR SAFETY AND UNDERSTANDING

Empathy is presented as essential for preventing harm. Shaw distinguishes between demonizing individuals and empathizing with the underlying causes of their actions. She coins ‘evil empathy’ to emphasize understanding without excusing wrongdoing, arguing that safety improves when we study psychology and social levers rather than relish in labeling. This stance informs how we interview offenders, design interventions, and craft policies to reduce risk while preserving human dignity.

MEMORY, MIRAGE OF TRUTH, AND DETECTION OF LIES

The discussion delves into memory distortion and false memories, highlighting how eyewitness testimony and witness statements can be unreliable. Shaw cites Elizabeth Loftus and other researchers on deception detection, noting professionals—police and investigators—often overestimate their ability to spot lies. The risk is wrongful conviction or acquittal. The takeaway is that memory is fallible, memory training and expert consultation are crucial in legal cases, and skepticism about confident gut judgments is warranted.

MURDER MOTIVES AND THE JUSTICE GAP

Shaw dismantles popular crime myths by describing murder as frequently impulsive, driven by immediate conflicts, debts, theft, or relationship dynamics rather than elaborate, premeditated plans. She emphasizes that recidivism for homicide is relatively low, and punishment models focused on deterrence may miss the mark. Restorative justice and family perspectives on forgiveness illustrate the complexity of healing and the potential for prevention when society attends to underlying causes rather than only punitive outcomes.

HEROIC IMAGINATION, INTERVENTION, AND BYSTANDER DYNAMICS

Drawing on the Stanford Prison Experiment and bystander literature, Shaw advocates for the heroic imagination: rehearsing how one would intervene in a crisis can increase real-world helping. She cautions that crowd dynamics and fear of standing out can hinder action, yet most bystanders do intervene in many real-world scenarios. The message is not to romanticize danger but to train minds to act when needed, thereby increasing safety and reducing the likelihood of passive complicity.

ONLINE DANGERS, INCELS, AND FRAUD IN THE DIGITAL AGE

Shaw highlights how entitlement and online echo chambers can radicalize or embolden harmful beliefs, exemplified by incel cultures. She also discusses fraud like the Tinder Swindler, where emotional manipulation exploits vulnerability and the desire for love. The modern risk is that technology—especially AI—can tailor deception, making it more persuasive. A core lesson is to cultivate healthy skepticism, verify claims, and recognize how social dynamics online can distort reality and precipitate real-world harm.

INTERVIEWING MONSTERS: ETHICS, NARRATIVES, AND RESPONSIBILITY

A practical thread runs through Shaw's work: interviewing those labeled evil requires ethical balance—curiosity about human motive paired with caution to avoid glamorization. She stresses that narratives reveal social patterns, not just sensational stories. The responsibility extends to producers and educators who shape public understanding. By prioritizing nuance, memory science, and the potential for prevention, interviewers can illuminate paths out of harm rather than merely documenting it.

Mentioned in This Episode

●Tools & Products

●Books

●People Referenced

Quantitative data points cited in the video

Data extracted from this episode

Metric	Population/Context	Value
Murder fantasies prevalence (men)	Two studies	Approximately 70%
Murder fantasies prevalence (women)	Two studies	More than 50%
Homicide recidivism	General population	1–3%
Vehicle NOx emissions (Dieselgate)	VW diesel cars in some cases	Up to 40x legal limit

Kinsey scale insights (brief distribution)

Data extracted from this episode

Aspect	Value	Source/Context
Men mid-range on Kinsey scale	About 50%	Kinsey scale discussion
Women mid-range on Kinsey scale	About 25%	Kinsey scale discussion

Common Questions

The dark tetrad groups four traits (psychopathy, narcissism, Machiavellianism, and sadism) on a spectrum. People can score low or high on each trait, and higher overall combinations increase the likelihood of harmful behavior. The relevant discussion starts around the segment explaining the four traits and their scales (timestamp ~126).