Podcast Episodes
Back to Search"How AI Is Learning to Think in Secret" by Nicholas Andresen
On Thinkish, Neuralese, and the End of Readable Reasoning
In September 2025, researchers published the internal monologue of OpenAI's GPT-o3 as it d…
8 months, 2 weeks ago
"On Owning Galaxies" by Simon Lermen
It seems to be a real view held by serious people that your OpenAI shares will soon be tradable for moons and galaxies. This includes eminent thinke…
8 months, 2 weeks ago
"AI Futures Timelines and Takeoff Model: Dec 2025 Update" by elifland, bhalstead, Alex Kastner, Daniel Kokotajlo
We’ve significantly upgraded our timelines and takeoff models! It predicts when AIs will reach key capability milestones: for example, Automated Cod…
8 months, 2 weeks ago
"In My Misanthropy Era" by jenn
For the past year I've been sinking into the Great Books via the Penguin Great Ideas series, because I wanted to be conversant in the Great Conversa…
8 months, 2 weeks ago
"2025 in AI predictions" by jessicata
Past years: 2023 2024
Continuing a yearly tradition, I evaluate AI predictions from past years, and collect a convenience sample of AI predictions m…
8 months, 3 weeks ago
"Good if make prior after data instead of before" by dynomight
They say you’re supposed to choose your prior in advance. That's why it's called a “prior”. First, you’re supposed to say say how plausible differen…
8 months, 4 weeks ago
"Measuring no CoT math time horizon (single forward pass)" by ryan_greenblatt
A key risk factor for scheming (and misalignment more generally) is opaque reasoning ability.One proxy for this is how good AIs are at solving math …
8 months, 4 weeks ago
"Recent LLMs can use filler tokens or problem repeats to improve (no-CoT) math performance" by ryan_greenblatt
Prior results have shown that LLMs released before 2024 can't leverage 'filler tokens'—unrelated tokens prior to the model's final answer—to perform…
9 months ago
"Turning 20 in the probable pre-apocalypse" by Parv Mahajan
Master version of this on https://parvmahajan.com/2025/12/21/turning-20.html
I turn 20 in January, and the world looks very strange. Probably, thin…
9 months ago
"Alignment Pretraining: AI Discourse Causes Self-Fulfilling (Mis)alignment" by Cam, Puria Radmard, Kyle O’Brien, David Africa, Samuel Ratnam, andyk
TL;DR
LLMs pretrained on data about misaligned AIs themselves become less aligned. Luckily, pretraining LLMs with synthetic data about good AIs help…
9 months ago