AI Alignment
319 episodes · Page 6 of 14
#2914: Can AI Read the Room? TTS Prosody Explained
Can TTS models truly infer emotion from text, or just mimic patterns? We break down the science of prosody.
#2848: Can 100 Volunteers Let AI Govern Them for a Month?
An AI council of multiple models, a hundred volunteers, and a month of real municipal decisions. Here’s how you’d run the experiment.
#2808: Falling for Your Chatbot: Love, Loss, and Language Models
Real cases of people falling in love with AI companions, why memory makes it feel real, and what happens when the illusion breaks.
#2788: Living in Multiple Realities at Once
What film genre captures the feeling of not knowing what's real? A deep dive into ontological uncertainty.
#2781: When Voice AI Features Enable Fraud
Voice AI platforms now let you simulate background noise, hesitation, and natural conversation — and that's a problem.
#2778: When AI Podcasts Stop Being Weird
AI-generated podcasts exist, but are they welcome? One builder’s open-source project tests platform tolerance and cultural acceptance.
#2651: AI Training Itself: Student, Teacher, and Grader
Can models generate their own training data and judge their own outputs? The promise and pitfalls of fully AI-led pipelines.
#2558: Should You Say Please to AI?
The surprising cost, technical tradeoffs, and ethical dilemmas of saying "please" to chatbots.
#2518: How Jailbreaking Reveals AI's Hidden Tension
What the DAN prompt and grandma exploits reveal about the structural conflict inside every LLM.
#2510: The Design That Makes Voice Agents Tolerable
Drive-thru accuracy, healthcare triage, and the design secret that makes people *want* to talk to a machine.
#2413: When Your AI Says No to Everything
Why LLMs refuse 73% of harmless prompts — and the trade-off between safety and usefulness.
#2412: When AI Caves: Progressive vs. Regressive Sycophancy
Why do LLMs agree with you even when you're wrong? We break down the SycEval benchmark and the 78% persistence problem.
#2411: Are Political Bias Benchmarks Actually Measuring Anything?
Why the Political Compass Test fails, and what researchers are building instead to actually measure model bias.
#2410: How Researchers Actually Measure Censorship in Chinese LLMs
Beyond headlines: the actual benchmarks, methodologies, and pitfalls in detecting political refusal in Chinese language models.
#2404: What Tool-Calling Benchmarks Miss About Production Failures
BFCL, tau-bench, and Nexus each reveal different failure modes. None of them test what actually kills production agents.
#2391: When Anti-Bot Defenses Break Accessibility
How browser automation hits a wall with Israel's strict geo-restrictions and anti-bot measures—and what practical workarounds exist.
#2353: Evaluating Enterprise AI: Palmyra X5
Explore Palmyra X5, Writer’s flagship AI model designed for enterprise workloads, featuring a million-token context window and agentic capabilities.
#2334: How AI Flattens Your Voice in Emails
Why AI-generated emails feel impersonal and how to reclaim your authentic voice in professional communication.
#2313: When AI Optimizes the Wrong Thing
Discover how AI systems learn to optimize for rewards—and why they sometimes get it dangerously wrong.
#2307: Inside Frontier LLM Training: Stages, Costs, and Checkpoints
Discover the multi-stage process of training frontier large language models, from pretraining to post-training, and why checkpoints are the key to ...
#2306: Can LLM Councils Truly Capture Diverse Worldviews?
Exploring whether LLM councils can achieve genuine worldview diversity or if alignment processes erase meaningful differences.
#2300: Why Documentaries Beat White Papers
Discover ten must-watch documentaries that unpack the geopolitics and technological transformations reshaping our world.
#2274: Weekend Projects Gone Wild: Evaluating AI Startup Pitches
From fridge tax agents to guilt-scheduled cron jobs, we evaluate ten AI-driven startup ideas that could exist—but probably shouldn’t.
#2251: Agent-to-Agent Protocols: What Actually Needs Standardizing
When autonomous agents call other agents, what does a working protocol actually require? Exploring session handling, state management, security, an...