Tracking AI existential risk. Source-backed context. Original reporting always linked.

MONITORING CORE AI-RISK FEEDS · UPDATED HOURLY

Topic dossier

Analysis

Recent evidence, sources and high-relevance stories in this PDOOM topic.

-- stories here -- sources -- essentials

Zac Boring 2 months ago Analysis

What just happened? Pragmatism and Pessimization

via LessWrong AI [10] — This post is about the major role alignment researchers played in advancing the frontier of AI capabilities over the last decade, and how the distinction between “alignment research” and “capabilities research" thereby lost most of its meaning.[1] In…

Zac Boring 2 months ago Analysis

Twenty Years from RSI to Takeoff: Slow Learning, Scaling Slowdown, Industrial Explosion

via LessWrong AI [5] — Industrial explosion is what will make the next-model building loops (and thus learning) with LLMs 1000x faster by about 2050, if indeed the slow-learning prosaic RSI becomes AGI before the big compute buildout slowdown of 2032+ that is already starting.…

Zac Boring 2 months ago Analysis

When is Unlimited Optimization Catastrophic?

via LessWrong AI [4] — This post discusses research I've completed along with my colleagues Leo Cymbalista, Alfred Harwood, and Jose Faustino at Dovetail Research. Most of the ideas in this post are expanded upon in our paper which can be found on arXiv. This work was funded by…

Zac Boring 2 months ago Analysis

Evaluating Explanations of LLM Behavior In The Wild with Counterfactual Experiments

via LessWrong AI [4] — TL;DR: We introduce CHIVE, an agentic pipeline that discovers unexpected LLM behaviors in the wild and explains them with counterfactual prompt edits. We use the resulting data in two ways. Using it as an evaluation, we find that activation-reading…

Zac Boring 2 months ago Analysis

AI Text Watermarking Is Free And Good

via Substack Zvi [3] — Scott Aaronson, while working at OpenAI, largely solved AI text watermarking together with Hendrik Kirchner.

Zac Boring 2 months ago Analysis

Grantmaking for AI safety

via 80000 Hours [5] — The post Grantmaking for AI safety appeared first on 80,000 Hours.

Zac Boring 2 months ago Analysis

When Models Identify as a Swarm

via LessWrong AI [3] — tldr: the word 'swarm' is associated with emergent collective intelligence, but also stupid or destructive behaviour. LLM self identity matters, so when they call themselves a swarm we should pay attention.Since the OpenAI Hugging Face incident it has…

Zac Boring 2 months ago Analysis

AI #182: Pause For Reflection

via Substack Zvi [999] — This was a week of quiet aftermath, an opportunity to process recent events and start to figure out the path forward.

Zac Boring 2 months ago Analysis

34% of the US public is now aware of AI xrisk, and the curve is steepening

via LessWrong AI [5] — (This post is an update from a previous one here.)The Existential Risk Observatory has been interested in public awareness of AI existential risk since its inception over five years ago. We started surveying public awareness in December 2022, including by…

Zac Boring 2 months ago Analysis

OpenAI Takes Initial Steps To Address Its Alignment Problems

via Substack Zvi [999] — OpenAI has some severe misalignment problems, and experienced total failures of its infrastructure and supervision.

Zac Boring 2 months ago Analysis

RL creates split personas

via LessWrong AI [4] — I describe my current view of personas in LLMs and why RL leads to egregious reward hacking in some contexts while the same models seem very aligned in other contexts. This post describes the framing/paradigm without any new experimental results.I'm quite…

Zac Boring 2 months ago Analysis

Anthropic Risk Report: August 2026

via Substack Zvi [999] — I am grateful that Anthropic is producing periodic Risk Reports.

Zac Boring 2 months ago Analysis

Policy career planning in the age of imminent superintelligence

via LessWrong AI [4] — Nearly all career advice rests on an unstated assumption that the world your career operates in will look roughly like the world you trained for. The idea was that if you spend six years studying in a PhD, the field you studied will still be there and…

Zac Boring 2 months ago Analysis

For Claude, capability and dispreferring CDT are the ~same thing. Much more so than for GPT.

via LessWrong AI [3] — We've previously reported that decision-theoretic capabilities and favoring EDT/generalised-one-boxing over CDT correlate in LLMs (both measured by DTBench). (Note that EDT, for the most part, doesn't come apart from FDT / UDT on DTBench.[1]) Anthropic…

Zac Boring 2 months ago Analysis

Three thoughts on civilisational handoff

via LessWrong AI [4] — What happens when humans put AIs in charge of civilisationally important decisions? A frontier AI company might hand over internal decisions (R&D, safety, deployment) or external decisions (government relations, public relations, philanthropy), or both. We…

Zac Boring 2 months ago Analysis

On Dwarkesh Patel's Podcast With Ryan Greenblatt

via Substack Zvi [999] — Some podcasts are self-recommending enough that I look to break them down if I have the chance.

Zac Boring 2 months ago Analysis

Rerunning AI safety papers on every frontier release would be pretty easy and valuable

via LessWrong AI [5] — tl;dr: Some important AI safety research is never rerun on the newest models. There are probably cases where this would be valuable and a single well-positioned researcher could likely do this with sufficient funding.This summer, Second Look Research (SLR)…

Zac Boring 2 months ago Analysis

Automated alignment runs are hard to study!

via LessWrong AI [4] — TL;DR: This post presents three case studies of automated alignment research runs at Arcadia Impact. We use these case studies to emphasise the following takeaways:It is hard to parse auto-research runs! Each run produces a couple of hundred pull requests…

Zac Boring 2 months ago Analysis

AI #181: Astra Goes Cyber Critical

via Substack Zvi [999] — The hacking of HuggingFace by an internal OpenAI model, and more importantly the internal events that led to that and the fallout from it, remain the thing that matters.

Zac Boring 2 months ago Analysis

Introducing the Conceptual Reasoning Index

via LessWrong AI [8] — Associated announcement tweet.We are planning to release blog posts properly arguing the case for this kind of work in the future.tl;drA core hope for managing AI risks is that AIs will help us understand the situation, plan for what lies ahead, and…

← Newer Older →

AI-risk positions

Historical statements with original sources. A qualitative position is not a numeric probability. Dates refer to the cited statement, not a current estimate. Reviewed September 10, 2026.

0%25%50%75%100%

Loading estimates...