Signatories to the EU's voluntary AI Code of Practice include Amazon, Anthropic, Google, Microsoft, and OpenAI, while Chinese AI companies remain absent
Nearly all dominant EU and US AI developers have signed up, but - so far - there are none from China.
Context & Ripple Effects
The sign-up list turns the EU’s newly published voluntary framework into a concrete alignment test for major model developers. The code was designed to help firms follow the AI Act through measures including current AI-feature documentation, as set out in the EU’s July code publication.
It also extends a pattern established by the earlier EU AI Pact, which drew Amazon, Google, Microsoft, OpenAI and more than 100 other companies. Google’s announced decision to join shortly before this list was published had already signaled momentum despite public resistance to the code from some companies.
First-order effects
- Amazon, Anthropic, Google, Microsoft and OpenAI are now publicly associated with the EU’s voluntary compliance route, including the code’s documentation-oriented expectations.
- The absence of Chinese AI companies leaves the current signatory group geographically uneven; the available coverage establishes non-participation, not the reasons behind it.
Second-order effects
- A shared voluntary framework gives participating developers a more common basis for preparing AI Act-facing governance and disclosures, while firms outside the group lack that public signal of alignment.
- The split makes the code’s uptake—not just its text—a test of the EU’s ability to convene global AI providers; Google’s decision to sign adds weight to the participating camp.
Third-order effects
- If major providers continue to use voluntary codes as an on-ramp to binding rules, compliance documentation and governance practices could become a competitive requirement for serving regulated markets.
- Uneven participation across jurisdictions could make AI governance increasingly shaped by regional blocs and state policy, though this signatory list alone does not show whether that divide will persist.
The trend: This is one data point in the shift from broad voluntary AI pledges toward regionally organized compliance regimes tied to formal regulation.