When the model resorts to unfaithful reasoning
This week: Biology of a LLM, entrepreneurs as theorists, a grammar of hypotheses
Topic
newsletter
A collection of 21 issues
If we tell them to maximize profit, they will
This week: Economic agents, cost of complexity, complexity comes and goes
A superhuman ability to bullshit
This week: Bullshit, a comprehensive survey, labour market disruption, and possibility spaces
Verification, backtracking, subgoal setting
This week: State anxiety, SimpleQA, self-improving reasoners (verification, backtracking, subgoal setting and something else…)
Feedback loops with the outside world
This week: Catastrophic risks, external validity, trust, machine assisted proofs
A narrative compelling enough to float
This week: Concentrated power, loopholes, AI pact, owners
The Unbearable Slowness of Being
This Week: Why do we live at 10 bits/s, the Anthropic economic index, our Icarus moment
At the edge of human reasoning
This week: Decoding-based regression, dataviz AI adoption, entrepreneurial resource mobilization, Knightian uncertainty
Self-Preference Your Way In Here
This week: Self-preferencing, a handy ChatGPT chart, exporting dictatorship tools, and brain dynamics
Competing with the platform
This week: Platform competition, professional autonomy, search fatigue