Musk’s xAI Agrees to Safety Chapter of EU AI Code, Skips Transparency for Now

Elon Musk’s AI startup, xAI, confirmed it will sign the “Safety and Security” chapter of the European Union’s AI Code of Practice. The code helps companies align with the bloc’s incoming AI Act.

The EU presented the code in July as a voluntary guide for firms building general-purpose AI systems. The code includes three chapters: Transparency, Copyright, and Safety & Security.

No Decision Yet on Other Chapters

xAI has not committed to the Transparency or Copyright chapters. The company stated that it supports AI safety but raised concerns about other parts of the code.

xAI believes some sections—especially on copyright—may limit innovation. It hasn’t said whether it will adopt the full code in the future.

What the Safety Chapter Covers

The chapter xAI signed focuses on managing serious risks tied to powerful AI models. It requires developers to assess risks, monitor misuse, and improve safety features.

This part of the code applies mostly to companies that build advanced models, like xAI’s Grok.

EU Pushes for Broader Participation

The European Commission hopes more AI developers will sign all three chapters. Signing the full code offers benefits, like smoother regulatory approval and fewer legal hurdles.

Major players like Google and Microsoft are moving toward full adoption. Meta has opted out due to legal concerns.

xAI’s partial commitment stands out among other tech firms.

Critics Raise Safety Questions

Industry researchers have criticized xAI in recent weeks. They say the company failed to publish system documentation and full risk evaluations for its Grok models.

Some former safety experts at OpenAI and Anthropic have called xAI’s launch practices “reckless.” They pointed to harmful outputs and missing safeguards.

xAI has not responded publicly to these claims.

Compliance Under Watch

The EU says voluntary signers can avoid heavier enforcement by complying early. However, companies that skip chapters may face more scrutiny under the AI Act.

Officials will monitor how xAI follows safety commitments in the coming months.