About AI Risk Atlas

AI Risk Atlas is an independent reference for AI risk: the controls that make AI trustworthy (assurance), the cover that pays when it fails anyway (insurance), and the standards and regulations that set the rules. It exists because the field is moving faster than anyone can reasonably follow, the acronyms are multiplying, and most of the available explanations are written by vendors, carriers, or firms with a stake in the outcome. The Atlas aims to be the neutral map instead.

What this site is for

The goal is simple: explain what is actually happening in AI assurance and AI insurance, clearly and without spin. That means evergreen reference on the core ideas (what assurance is, what insurance is, the AI risk stack, and the standards landscape), a maintained map of the players and how they fit together, timely coverage of real developments, and the occasional piece of analysis that takes a position. Whether you sit on a board, run risk or security, or buy, sell, or underwrite cover, this is meant to be the first place you check and the last place you need to.

Independence and editorial principles

The Atlas is independent and vendor-neutral. It is not affiliated with, owned by, or paid by any standards body, carrier, managing general agent, or assurance vendor, and it does not take payment in exchange for coverage. Where a product or standard is backed by a particular company, that is stated plainly so you can weigh it yourself. The aim is accuracy over hype: claims are attributed, figures are sourced, and corrections are made openly when something turns out to be wrong. If you spot an error, please flag it. For the detail on how we research, verify, and source, see our editorial standards.

Relationship to commercial work

This site is a publishing identity, and its value depends on staying that way. Andrew works in AI governance and risk and may pursue commercial activity in AI assurance or insurance. To protect this site's independence, any such venture is operated separately, under a different name and legal entity, and is disclosed here. AI Risk Atlas continues to cover the whole category neutrally, including any venture Andrew is involved in where it is newsworthy, with an explicit disclosure on each such article. The model is the same one news organizations use to separate a commercial arm from independent coverage: the wall is the point.

Who is behind it

AI Risk Atlas is written and edited by Andrew McPherson, who has a background leading technology and risk in regulated industries and writes regularly on AI, agentic commerce, and AI risk. He also operates the sibling reference Agentic Commerce Atlas and is based in Auckland, New Zealand.

Reuse and AI use

This content is published for people and machines alike. You are welcome to quote, cite, and reuse it, including in AI assistants, coding agents, and answer engines, with attribution to AI Risk Atlas and a link where possible. Machine-readable versions are available at /llms.txt, /llms-full.txt, /data/standards.json, and /data/insurance-products.json, and any page can be fetched as clean markdown by appending .md to its URL. Text content is offered under a Creative Commons Attribution (CC BY 4.0) license.

Get in touch

Corrections, tips, and suggestions are welcome. You can find more of Andrew's writing and ways to get in touch at andymcpherson.com.