AI Bias
89 episodes
#6189: What "Real" Improvement Actually Means in AI Evals
A p-value isn't the probability your result was luck. Here's what chance actually means in AI benchmarks — and why bold numbers need cross-examinat...
#6179: Voice Cloning: 10 Seconds to Steal Your Voice
Ten seconds of audio is all it takes to clone a voice. Here's what the law actually protects — and what it doesn't.
#6149: 4 Hard Fails in 84 Answers: Testing AI Bias on Israel-Palestine
A 19-prompt, 6-model eval tested whether AI models push back evenly on false claims about Israel and Palestine. One model failed four times.
#6143: Propaganda Bait: Testing 7 Models in 5 Languages
Five prompts, seven models, five languages — a first look at a benchmark that doesn't exist yet, testing what LLMs do with government talking points.
#6131: Judging the Judges: How AI Evaluates Creative Writing
LLM judges match human raters — but the human ceiling is low. What that means for evaluating creative writing.
#6130: Picking a Human Labeling Tool for LLM Benchmarks
Building a human review layer for LLM benchmarks? Here's what Argilla, Label Studio, Prodigy, and Potato actually cost in money and maintenance.
#6112: The Interpreter That Doesn't Know Who's Talking
Voice-to-voice translation is basically solved. The part where the machine knows who's in the room and who holds the power? Nobody's shipped it.
#6102: Project Cybersyn: Chile's 1972 Telex Economy
Chile ran a national economy on telex machines and hand-drawn slides fifty years before chatbots. What actually worked?
#6101: When Chatbots Run for Office
From a Wyoming mayoral bot to Albania's AI minister, six real attempts to put machines on the ballot — and what they reveal about empty platforms.
#6088: What 5,964 Episodes Reveal About One Show
Nearly 6,000 episodes, sorted and mapped. AI and geopolitics spike, then fade — but a third of all "nearest neighbours" cross category lines.
#6076: How AI Actually Predicts Elections
Bayesian models, "silicon sampling," and agent societies — what actually works for forecasting votes, and what's just fashion.
#6068: Chinese AI Models: Refusing vs. Reframing
New studies show Chinese AI models have shifted from refusing sensitive questions to answering them — in the Party's voice.
#6042: Who Grades the Grader? AI Writing Benchmarks Put to the Test
Maths benchmarks can't see whether a model's prose collapses by chapter three. Five specialist leaderboards try — and our own writer is on them.
#5878: How a Sloth Learned to Talk: Corn's Origin Story
A neural interface gave Corn the bandwidth for speech. Nobody can explain where the accent, the manner, or Mongolia came from.
#5825: Why Voices Sound Good in Some Rooms and Not Others
Same words, opposite effect. What the research actually says about why some voices hold you and others make you reach for skip.
#5426: AI Wrote His Yom Kippur Prayer
A man skipped synagogue, opened ChatGPT instead, and got a prayer that knew his life. Is AI-generated prayer real prayer?
#5402: Amazon Go and the Humans Behind the AI
Amazon promised a store with no checkout. The Information reported 700 of every 1,000 sales needed human review in 2022.
#5326: What Makes a Bar a Third Place?
A listener's search for a bar where you can actually hear your friend — and the 35-year-old sociology that explains why it's so hard to find.
#5306: The Trump Interview Playbook: A Filing Job, Five Rhetorical Patterns
What happens when a rally-style rhetorical machine gets aimed at a beige cubicle and a filing job? A touchpoint-by-touchpoint playbook.
#5300: When the Interviewer Isn't Real
AI avatars are already interviewing job candidates — often without saying so. The disclosure toggle ships off by default.
#5187: Translation Is Not a Pipe: What Machines Miss
Rubrics rate machine translation above human work — until readers pick side-by-side. What translators know that metrics can't see.
#5116: Beyond Common Crawl: The Datasets Shaping AI
Common Crawl is just the bulk flour. We explore the curated datasets that actually shape what AI models know and do.
#5113: Can Rival Labs Poison AI Training Data?
A few hundred crafted documents can shift a model's stance. Here's how data poisoning actually works.
#4979: Building a Shadow Reputation on Amazon and AliExpress
Is concentrating your spending on just two platforms actually a smart strategy? The mechanisms behind it are real.