Tristan Harris opens by zooming out from misinformation to the broader, cumulative effects of technology on human minds and institutions. Framing with E. O. Wilson’s line about Paleolithic emotions, medieval institutions, and godlike technology, he argues our brains and governance are outpaced by AI systems. He reframes social media as humanity’s first contact with AI: every swipe activates a supercomputer that predicts the next perfect stimulus from billions of data points. That first contact produced information overload, doom-scrolling, loneliness, polarization, deepfakes, and a breakdown of shared reality. While platforms promised connection and relevance, the engagement business model rewarded addiction to attention and instant virality. Design choices like pull-to-refresh, likes, beautification filters, and one-click resharing created a race to inflate status and amplify extremes.

With large language models, he warns of a looming second contact before the first is fixed. AI will steepen the complexity of existing risks, from misinformation to synthetic biology. Concrete examples include viral AI-generated hoaxes, like the fake Pentagon explosion image that briefly rattled markets, illustrating how cheaply trust can be hacked at scale. Solutions, he argues, require more than content moderation. We need a wiser information ecosystem and upgraded institutions that handle diffuse, long-horizon harms, not just acute incidents.

His roadmap: embrace how human brains really work by designing for belonging and offline community, upgrade institutions to govern chronic harms and set liability that matches systemic risks, and bind race dynamics so companies do not win by externalizing harm. You cannot wield godlike powers without the wisdom, love, and prudence of gods, he concludes. The call is to realign technology, incentives, and governance so sense-making and choice-making keep pace with accelerating power.

Tristan Harris opens by naming the root of scandals and grievances in tech: an extractive attention economy that hijacks human frailties. Using E. O. Wilson’s line about “Paleolithic emotions, medieval institutions, and godlike technology,” he frames the problem as not machines overwhelming our strengths, but our weaknesses. Drawing on his background as a magician and Google design ethicist, Harris explains how persuasive design exploits attention, validation-seeking, and social proof. Examples include slot-machine mechanics in phones and apps, YouTube’s recommender system keeping users in trance states, and Facebook group recommendations fueling conspiracy ecosystems.

He introduces the concept of human downgrading: the systemic erosion of attention spans, free will, mental health, civility, and trust. Evidence includes teen depression trends, polarization metrics, and how recommender AIs tilt content toward extremity (“Crazytown”) to maximize engagement. Harris shows how platforms built predictive “voodoo dolls” of users, making persuasion inevitable without needing microphones or data theft. Deepfakes and synthetic media amplify the problem, overwhelming human judgment.

Band-aid fixes like grayscale, notifications, blockchain, or ethics training are insufficient. The solution requires systemic change: shifting from artificial to humane social systems, from overwhelming AI to fiduciary AI aligned with human limits, and from extractive to regenerative incentives. He demonstrates how humane design can wrap around human physiology, cognition, and social needs—such as designing for trust, belonging, and small-group conversation rather than polarization.

Harris closes with urgency and hope. Just as shared language like “time well spent” spurred Apple, Google, and YouTube to adopt well-being features, a new shared agenda can spark a “race to the top” in humane tech. He urges leaders, workers, journalists, policymakers, investors, and entrepreneurs to unite behind this civilizational moment: to embrace our Paleolithic emotions, upgrade medieval institutions, and wield godlike technology with wisdom.

The trailer opens by revealing that Google’s autocomplete and platform feeds differ by user, location, and inferred interests. It reframes this as a design choice, not an accident. Former leaders and designers from Facebook, Pinterest, Google, Twitter, and Instagram explain that product teams intentionally leverage psychology to maximize engagement. The montage links hearts, likes, and thumbs up to shallow validation and misinformation, while on-camera experts describe measurable harms: anxiety and depression, polarization, and loss of shared reality.

Key claims highlight that social platforms can influence offline behavior without user awareness and that falsehoods travel faster than truth. The voiceover warns that when everyone has “their own facts,” society fragments. The trailer escalates to a governance-level warning: if you wanted to control a population, a platform like Facebook is an unprecedented tool. It closes with the filmmakers’ thesis that creators have a responsibility to change course and that failing to rebalance incentives could be “checkmate on humanity.”

Trailer promoted by Netflix as the official preview for the 2020 feature documentary.

Harris opens with a personal reflection on time slipping away through compulsive checking of email, feeds, and notifications. He compares phones and apps to slot machines, engineered with variable rewards that exploit human psychology. Even knowing this, he admits it’s hard to resist.

He highlights the cost of constant interruptions: research shows each distraction takes ~23 minutes to recover from, and frequent external pings condition us to self-interrupt. Harris demonstrates how design could restore choice—for example, chat systems that let someone “focus” while still allowing urgent messages through.

He urges designers to upgrade goals from surface-level metrics (e.g., ease of sending a message) to deeper human values (e.g., quality communication). Examples include Couchsurfing, which measured its success in “net positive hours” of meaningful experiences created.

Harris imagines a broader system: social networks tracking meaningful connections, dating apps optimizing for fulfilling relationships, and professional networks focusing on job outcomes—not just engagement. He suggests labels like “organic” or LEED certification could inspire a new category of humane tech, prioritized in app stores and browsers.

He concludes with a call to action: leaders must adopt new metrics, designers should embrace a Hippocratic oath for design, and users must demand technology that contributes positively to human life. The shift would move us from a world optimized for time spent to one focused on time well spent.