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

The Psychology of "Evil" and Human Nature

• Murder fantasies are extremely common, with studies indicating approximately 70% of men and over 50% of women have fantasized about killing someone.
• Julia Shaw argues against a binary "monster vs. non-monster" classification, positing that dark traits (psychopathy, sadism, narcissism, Machiavellianism) exist on a continuum where everyone scores somewhere.
• Shaw rejects the concept of being "born evil," citing Adolf Hitler's early life as an example of how harmful behaviors often manifest over time through socialization rather than innate nature.
• The term "evil" is criticized as a conversation-stopper that dehumanizes perpetrators, preventing the understanding necessary to prevent future crimes.
• Large-scale harm requires two psychological mechanisms: dehumanization (collapsing empathy for the "other") and de-individuation (losing individual identity within a group).

Deception, Memory, and Criminal Justice

• Police officers and laypeople are notoriously bad at detecting lies; studies show they perform at chance levels despite high confidence in their "intuition."
• False memory research demonstrates that up to 70% of participants can be convinced via suggestive interviewing they committed a crime that never occurred.
• Memories are constructed and recombined; "gist memory" (the general event) is usually accurate, while "verbatim details" (specifics) are highly prone to distortion and implantation.
• Collective memory consolidation (discussing events together) can increase the quantity of recalled details but significantly increases the risk of weaving a shared false narrative if not recorded individually first.
• The criminal justice system often misallocates resources by focusing heavily on low-recidivism violent crimes (murder) while neglecting high-recidivism fraud and elder abuse.
• Recidivism for homicide is extremely low (1–3%), suggesting that long prison sentences for murderers may be less effective for public safety than targeted prevention for other crime types.

Relationships, Sexuality, and Bias

• Bisexuality is often misunderstood as a "phase" or a stepping stone to being gay, despite historical data showing it was more common among men prior to social shifts in the 1970s.
• Jealousy is identified as a red flag for insecurity and control rather than a legitimate sign of love, and research suggests monogamy is a social construct that often leads to infidelity.
• The "Tinder Swindler" case illustrates how fraud succeeds by exploiting the human desire to believe in a narrative that confirms our deepest emotional needs (love and security).
• In relationships, partners often hold conflicting memories of shared labor (e.g., one claims 60% of chores, the other 50%), highlighting the subjective nature of personal accounts.
• Bisexual individuals face unique stigma, including hyper-sexualization by men and the "not gay enough" or "actually gay" dismissal from the LGBTQ+ community.

Environmental Crime and Future Directions

• "Green Crime" (illegal poaching, corporate fraud, pollution) is driven by similar human psychological factors as violent crime, including conformity, rationalization, and the normalization of unethical behavior within an industry.
• The Volkswagen "Dieselgate" scandal involved engineers who knew their defeat devices emitted 40 times the legal limit of nitrous oxides but concealed the data due to market pressure and conformity.
• Environmental crimes are often committed by "ordinary" people in groups, utilizing shell companies and sophisticated deception to bypass regulations, rather than by individuals with overt psychopathic traits.
• Shaw is optimistic about technology's potential to aid justice, specifically through AI tools designed to administer the "cognitive interview" without leading questions, thereby reducing false memory creation.
• Shaw founded "Spot," a platform using AI to record contemporaneous accounts of workplace incidents or emotional events to preserve accurate memory before it can be distorted by social interaction.
• Future risks include AI systems acting as "false memory machines" that tailor narratives to user desires, potentially creating a feedback loop where users' memories of the past are reshaped by AI interactions.