6 newsletters, ranked by editorial quality and how recently they published. Looking for something more specific? Search instead.
A blog on economic history that critically re-examines prominent empirical claims in economics, such as the effects of the French Revolution on growth and the 'China shock'. Readers get close readings of papers, with attention to methodology and quotations from the original sources.
Pre-registered, publicly documented ML experiments with specific numbers, covering LLM-as-a-judge reliability, API versus local model costs, and EWC versus flat L2 regularisation. Readers get a transparent look at what each experiment tested, whether its gates passed or failed, and why.
Essays on AI governance that define concepts rigorously, such as coherence as total correlation, and dissect real incidents and studies. Readers get critical analysis of agentic AI failures, accountability gaps, and how research findings get misrepresented.
AI Impacts is a research group whose blog reports on its Expert Survey on Progress in AI, covering survey results, methodology explanations and reanalyses of expert forecasts about advanced AI. Readers get concrete survey figures on researcher opinion about AI risk and timelines, plus transparent discussion of how such surveys are conducted.
A systems-minded newsletter on long-horizon infrastructure thinking drawn from distributed systems experience, covering durability and change that survives. Recent issues cover AI sycophancy in engineering teams and a structured, evidence-driven research methodology for AI-assisted analysis.
Short opinion-style essays on scientific research topics, covering public health data, research reproducibility and AI in mathematics. Each issue builds a commentary around a recent study or news item in science.
One short email a week with hand-picked newsletters from the index. Free, no spam, unsubscribe anytime.
