AI Freaky Facts: The AI Documentary

Automated license plate cameras now perform twenty billion scans a month across six thousand U.S. communities. This episode investigates Flock Safety, the Fourth Circuit fight over warrantless tracking, and how ICE accessed school cameras without local consent. From a Texas funding freeze to citizens cutting down poles, the backlash is reshaping surveillance policy nationwide. Should artificial intelligence track every driver witho...

AI accused hundreds of thousands of people of fraud without a single human review. Michigan's MiDAS system wrongly flagged nearly forty thousand unemployment claims, while Australia's Robodebt scheme used flawed income averaging to generate false debts tied to severe hardship and multi-billion-dollar settlements. This episode investigates how automated fraud detection erased the presumption of innocence, and why no one checked the ...

AI is inventing court cases that never existed, complete with fake judges, docket numbers, and quotations, and lawyers are still filing them. This episode follows the ChatGPT case that started the pattern, the Nebraska attorney suspended after citing invented precedent, and the UK ruling that fake citations could amount to perverting justice. It examines why even legal specific tools still hallucinate, and the verification habits n...

AI is already coordinating rent across competing landlords, producing the outcome of a conspiracy without a single phone call. This episode follows the Department of Justice case against RealPage, the Seattle block where ten managers relied on one shared pricing engine, and the hub and spoke system that pooled nonpublic competitor data. It examines the agreement gap in antitrust law and the new California, New York, and city level ...

AI and brain implants now let paralyzed patients type, draw, speak, and play games using thought alone. This episode follows real Neuralink and Synchron trial participants who regained independence through brain computer interfaces, and examines the hidden risks behind that progress. It explores remote neural signal hijacking, the limits of informed consent, and the danger of abandoned medical hardware. Can artificial intelligence ...

AI is already generating quotes that were never spoken and bylines for reporters who never existed. This episode follows how a fabricated quotation reached the New York Times, why Sports Illustrated ran stories under AI created authors, and how a fictional freelancer fooled six major publications for months. It reveals the newsroom economics that let verification break, and the emerging tools now restoring transparency, from conten...

AI reads your medical scans before your doctor does, and no patient is told. A mock trial shows jurors judge radiologists mainly on the order they viewed an AI result, not the medical decision itself. A newly authorized breast cancer risk model predicts disease five years before any tumor appears. Doctors who rely too heavily on AI can lose their diagnostic edge, and consent forms quietly omit the algorithm shaping your diagnosis. ...

AI powered humanoid robots are stepping onto factory floors with no safety cage between their bodies and the people working beside them. This episode follows a fifty one million dollar lawsuit after a Tesla robot injury, and the Korean union that forced Hyundai to negotiate labor terms before deploying twenty five thousand Atlas units. From Boston Dynamics to Figure AI, humanoid robots are entering real workplaces faster than regul...

AI is no longer just changing what we do. It predicts when we will quit, recreates us after death, denies our healthcare in one point two seconds, clones our voice from three seconds of audio, and rewires how our children think. This is Episode fifty of AI Freaky Facts, and it steps back from any single story to ask what all of them, taken together, are actually telling us. What does it mean to be human in a world where the most co...

AI cloned a CFO, his colleagues, and ran an entire video meeting in real time. Nobody on the call was real. A finance worker transferred twenty five million dollars before anyone realized what had happened. Deepfake fraud attempts rose thirteen hundred percent in a single year. Voice cloning now needs only three seconds of audio. This episode investigates the deepfake fraud epidemic, how attackers harvest public footage, why organi...

AI is deciding whether the orders from your doctor get paid for, and it is getting them wrong at a rate that should alarm everyone. Cigna algorithm processed three hundred thousand claim denials in two months at one point two seconds per claim. UnitedHealthcare system has a 90% reversal rate on appeal. Medicare just launched its own AI denial pilot. This episode investigates the algorithm behind your health insurance, who built it,...

AI brought a dead man back to speak in court, his face and voice recreated from old footage, saying words he never recorded. As griefbots, voice clones, and digital avatars spread, startups are building chatbots from the texts, voicemails, and posts people leave behind. The digital afterlife industry is projected to reach nearly eighty billion dollars within a decade. This episode investigates who controls a digital life after deat...

AI is scoring your likelihood of quitting before you feel it coming. Predictive HR systems analyze emails, calendars, performance data, and behavioral signals to label workers as flight risks in secret. Bias and mental health profiling shape careers without consent. In this episode, Steve exposes the predictive HR machine. What happens when artificial intelligence predicts your future at work?

Episode notes at: https://aifreakyfacts...

AI algorithms are profiling the emotional vulnerabilities of teenagers and using them to maximize engagement. Amnesty International found that TikTok AI served suicidal content to teen test accounts within forty five minutes. A French family says an AI recommendation algorithm pushed their daughter toward suicide. Australia became the first country to ban social media for under sixteens. What is artificial intelligence doing to the...

AI facial recognition has turned your face into a product. Clearview AI scraped thirty billion photos without consent to build the largest face database in history. Retailers are scanning shoppers without disclosure. Meta has paid more than two billion dollars in settlements for capturing facial geometry without permission. Steve Atwal investigates how your biometric data is being collected, commodified, and sold, and whether the l...

AI translation tools are making life-altering mistakes in asylum interviews, hospitals, and courtrooms, and the people harmed rarely know it happened. A woman fleeing abuse was denied asylum when an app mistranslated "father" as "boss." A Syrian mother underwent surgery without informed consent. A BBC investigation linked mistranslation to more than eighty infant deaths in UK hospitals. Steve Atwal investigates how artificial intel...

AI bossware is logging your keystrokes, screenshots, and activity, often without your knowledge. Steve Atwal investigates how workplace surveillance spread from banks to warehouses, why monitored workers quit at nearly double the rate, and why the smartest companies are choosing trust over tracking. Will workers inside these systems demand something better before artificial intelligence algorithms decide that asking the question is...

AI is already deciding who gets seen first in emergency rooms, and who waits. Across hundreds of hospitals, triage algorithms have missed real sepsis cases, under‑referred Black patients, and blocked chronic pain prescriptions with no appeal. As transparency rules are rolled back, who is accountable when artificial intelligence systems quietly decide you can wait?

Episode notes at: https://aifreakyfacts.com/stories/

Topics Covered:

AI...

AI agents are making corporate decisions without human approval. A company collapses. The board blames the CEO. The CEO blames the software that has been quietly running the business for 18 months. Sixty percent of executives now rely on AI for decisions, but only five percent manage it effectively. In 2026, California killed the "autonomous harm" defense, Colorado mandated impact assessments, and the EU AI Act introduced personal ...