Cheatsheets home

AI cheat sheets

This is the section of the collection I use most at work. The frontier lab comparison, the model picker and the API pricing table are the pages I keep open while building agentic systems, and they get re-verified as the labs ship new models.

The second half is the safety side: a field guide to existential risk, a timeline of the alignment problem, a directory of the safety ecosystem, and an interactive p(doom) calculator that shows which assumptions actually move the estimate. The two halves belong together. You cannot reason about the risk without knowing what the labs are shipping, and you cannot pick a model responsibly without knowing its failure modes.

Start with the frontier labs page if you want the landscape, the model picker if you have a task in hand, and the agentic AI cheat sheet if you are building.

22 references filed under AI & Safety. Most read right now: 8 Frontier AI Labs Compared (2026): OpenAI vs Anthropic; AI Coding Agents Compared: Claude Code, Cursor, Copilot; AI Existential Risk Cheatsheet: AGI, Safety & Mitigation.

Start here

  1. 8 Frontier AI Labs Compared (2026): OpenAI vs AnthropicThe landscape in one page: eight labs, ranked and compared
  2. Which AI Model Should I Use? Anthropic & OpenAI Model PickerWhich Anthropic or OpenAI model for the task in front of you
  3. Agentic AI Cheatsheet: Patterns, Failures, Autonomy LadderArchitecture patterns and failure modes for systems that act

Press ⌘K or / to search section headings.

205 references · 15 fields · 1977 sections indexed · 1541 cross-links Last change: wordpress-to-static: reflect the fleet's move to Cloudflare… just now Reviewed this week: 12 2,389 views/day Trending: 8 Frontier AI Labs Compared… · Starlink Satellite Anatomy… · AI Coding Agents Compared…

Every sheet here is produced by a governed Claude Code pipeline: a version-controlled spec as acceptance criteria, primary-source research, a self-verification gate, and a public git audit trail. Read how it is built or browse the source on GitHub.

22 of 205 AI & Safety×
AI & Safety

AI Progress Live Dashboard

Dashboard tracking AI progress across four benchmarks (SWE-bench Verified, METR time horizons…

VisualJun 10, 2026 · upd Sep 27, 2026Open
AI & Safety

p(doom) Calculator Test Harness

Internal test harness comparing p(doom) calculator formula variants (multiplicative, weighted…

ReferenceNov 23, 2025 · upd Sep 18, 2026Open

The map draws itself with JavaScript, which is off. The grid below lists every sheet, and each sheet's detail view names its neighbours.

Drag to pan, scroll to zoom, double-click to reset. Click a node to open its detail. Colour is category; size is how much it is read this month.

Get new references and build notes

Occasional email when a new reference ships or the pipeline changes. No spam, no tracking, unsubscribe anytime.

↑↓ moveenter openg then g/m/p lensesc close

Keyboard

⌘K / Ctrl K / /
Open the search palette
↑ ↓ enter
Move and open inside results
esc
Close the palette or the drawer
g then g
Grid lens
g then m
Map lens
g then p
Paths
t
Toggle theme
?
This map