Podcast Episodes
Back to Search
Can Anyone Still Spot an AI Voice?
Episode 6215
Two Chatterbox hosts are talking to you right now, and this audio carries an inaudible watermark — can you hear it? That is the design goal, and it i…
1 week, 2 days ago
Nobody Trains an Uncensored Model From Scratch
Episode 6214
There's a shelf of uncensored models, but it's stocked with fine-tunes of other labs' bases, not from-scratch builds. This episode surveys what actua…
1 week, 2 days ago
Ten Weirdest AI Evals Ever Published
Episode 6213
What happens when researchers need to measure machine intelligence and end up grading puns, emoji, and a joke with the punchline surgically removed? …
1 week, 2 days ago
Woolly Sedate Snail and Other Model Failures
Episode 6212
Hugging Face hosts nearly three million model repos, and 85.6% have under 200 downloads. This episode wanders the long tail of AI: auto-generated "wo…
1 week, 2 days ago
Attention Is Still Winning: The Hybrid Model Census
Episode 6211
Nine months after our quadratic bottleneck episode, the post-transformer world has actually shipped — just not in pure form. This episode is a census…
1 week, 2 days ago
Who Decides If an Election Is Real?
Episode 6210
There's no office anywhere that stamps an election genuine, yet most people assume one exists. This episode digs into the numbers behind sham electio…
1 week, 2 days ago
Watching Elections: Who Collects the Numbers, and Why
Episode 6209
Elections are the one moment a government publishes its intentions in advance — which is exactly why foreign services, embassies, and trading desks w…
1 week, 2 days ago
Who Watches Elections, and Who Buys the Numbers?
Episode 6209
In any big democratic election, some people care intensely about the process rather than the winner. This episode separates watching an election from…
1 week, 2 days ago
Israel's 120-Seat Forecasting Ministry
Episode 6208
Israel elects 120 seats by pure proportional representation, nobody wins a majority, and a party at 3.25% gets nothing at all. So what would it take …
1 week, 2 days ago
Two WERs, One Model: Inside ASR Eval Harnesses
Episode 6207
You fine-tune a Whisper-style model or Parakeet on your own voice, run it twice, and get two word error rates several points apart. Neither number is…
1 week, 2 days ago