Skip to content

About

What this is

Section titled “What this is”

Comparative AI is an independent academic research project with a narrow, explicit aim:

to build a topic-first, source-grounded, traceable, and contestable comparative atlas of AI governance across three jurisdictions — China, the United States, and the European Union.

Why

Section titled “Why”

Excellent overview literature on AI governance already exists in both Chinese and English. What remains scarce, especially as a freely accessible resource, is a body of work that combines the following three features:

  1. Horizontal, topic-by-topic comparison (rather than country-by-country surveys).
  2. Full traceability to primary texts — every claim on the site can be clicked through to the original document.
  3. Treating hard law, soft law, and corporate practice as three interrelated families of material, organised together. Frontier model behaviour, for instance, is often nowhere to be found in executive orders — it lives in Anthropic’s Responsible Scaling Policy. Voluntary industry standards, conversely, sometimes end up written into government guidance.

Who it is for

Section titled “Who it is for”
  • Policy researchers comparing across jurisdictions, who want to save time on primary-source retrieval.
  • Compliance practitioners mapping statutory text onto concrete product decisions, who need a cross-checking starting point.
  • Product managers and engineers trying to understand where company policy ends and external regulation begins.
  • Students and newcomers seeking the skeleton of the field.

This site is not journalistic commentary for a general audience, nor is it legal advice from an attorney.

What this is not

Section titled “What this is not”
  • Not legal advice. For any concrete compliance question, please consult a practising lawyer.
  • Not a news aggregator. We do not chase breaking stories or write reactive commentary.
  • Not advocacy. The goal of comparison is to surface structural differences, not to rank systems as better or worse.
  • Not a machine-translation product. All translations are human-reviewed; AI-assisted passages are explicitly marked.

Author

Section titled “Author”

The site is maintained by an independent researcher working in the AI / LLM industry. Personal details and contact information are deliberately minimal. For formal collaboration or complaints, please use GitHub Issues.

Next steps

Section titled “Next steps”