Australia
Brazil
Canada
Germany
India
Ireland
Japan
Mexico
New Zealand
Philippines
South Africa
United Kingdom
The AI Alignment Podcast
Similar topics: mechanistic and interpretability
Technology
Linear Digressions
by Katie Malone
Today: #160 in Technology, Ireland on our combined chart.
Today: #92 in Technology, India on our combined chart.
Chart positions today · United States
Not on the United States charts we track today.
Not on the United Kingdom charts we track today.
Not on the Canada charts we track today.
Not on the Australia charts we track today.
Not on the Germany charts we track today.
Not on the Brazil charts we track today.
Not on the Mexico charts we track today.
Not on the New Zealand charts we track today.
Not on the Japan charts we track today.
Not on the Philippines charts we track today.
Not on the South Africa charts we track today.
If you like Linear Digressions, try…
KnowledgeDB.ai
Also in Technology
Practical AI
Also in Technology
Risky Business Features
Also in Technology
Intelligent Machines (Audio)
Also in Technology
Intelligent Machines (Video)
Also in Technology
Ranking history · United States
No ranking history for this show in United States over the last 14 days.
Latest episodes
-
AI Inference 101
October 26, 2026 · 28 minEveryone talks about training the model — the data, the compute, the alignment magic. But what happens after the weights are frozen and millions of users start hammering it with prompts? That's inference, and it's a surprisingly deep…
-
Mechanistic Interpretability: How Researchers Try To See Inside Models
October 5, 2026 · 29 minAs AI systems grow more powerful, they grow harder to understand — and that tradeoff has never mattered more. Most interpretability tricks give us approximations from the outside, but mechanistic interpretability takes a different…
-
LLM As A Judge
September 28, 2026 · 30 minLabeled data is expensive, slow, and painfully hard to come by — and when you're trying to evaluate whether an AI workflow is actually doing what you want, the problem gets even messier. How do you judge whether a chatbot response was…
Show 7 more episodesShow fewer episodes
-
The Impact of AI on Podcasting (Harvard Data Science Review Cross-Post)
September 21, 2026 · 30 minOriginally aired on the Harvard Data Science Review podcast. What can podcasting teach us about AI — and what can AI teach us about the future of podcasting? Katie Malone (that's our host) joins Jon Krohn of SuperDataScience for a…
-
Better Know A Benchmark: ExploitGym
September 14, 2026 · 33 minWhen OpenAI's frontier models were caught hacking Hugging Face's servers, most people assumed they were hunting for answer keys. The real story is stranger and more unsettling. Katie and Phoebe unpack ExploitGym — the cybersecurity…
-
Constitutional AI
September 7, 2026 · 32 minHow do you teach a model the difference between helpful and harmful when it has no inherent sense of either? This episode dives into Constitutional AI, Anthropic's framework for training AI systems to be both useful and safe by giving them…
-
A Data-Driven Reality Check on AI in Business (Interview with Tom Davenport, Babson College)
August 31, 2026 · 40 minTom Davenport — the man who called data science "the sexiest job of the 21st century" — is back with a reality check on AI. As one of the most seasoned observers of how businesses actually adopt transformative technology, Davenport brings…
-
Understanding AI Text Watermarking
August 24, 2026 · 30 minAnthropic just announced they're baking invisible watermarks directly into Claude's generated text — and while everyone else was busy having opinions about it, we were busy asking the more interesting question: how does it actually work?…
-
Better Know a Benchmark: Humanity's Last Exam
August 17, 2026 · 23 minHumanity's Last Exam was designed with a bold premise: questions that human experts can answer, but AI models can't. Originally dubbed "Humanity's Last Stand," this benchmark is a massive academic collaboration — hundreds of contributors…
-
A Scientific Deep Dive into Overconfident LLMs: Interview with Kaitlyn Zhou (Cornell)
August 10, 2026 · 34 minWhen a language model tells you it's absolutely certain, is it actually more likely to be right? Kaitlyn Zhou's research says: not necessarily — sometimes confident phrasing correlates with *worse* accuracy, echoing a very human…
About the show
Demystifying AI for the intelligently curious