Germany: Pausing AI Development Not Feasible for Europe, Urges Global Governance Cooperation

On September 14, Germany's digital affairs ministry responded to calls from leading AI developers to slow down model iteration and strengthen safety guardrails, saying that halting research and development is not a viable option for Europe. A ministry spokesperson said that continuing AI research and innovation is essential to consolidating Europe's digital sovereignty.

The response follows a long post published over the weekend by Dario Amodei, CEO of Anthropic, urging AI companies to slow the pace of model capability advancement and proposing stricter safety mechanisms, including independent assessors embedded within AI firms. The German ministry said it is closely monitoring global AI developments and takes the public risk warnings from top AI researchers seriously.

The spokesperson also noted that if AI systems can operate outside test environments and connect to other systems on their own, this would create "a fundamentally new threat scenario" requiring new policies. He expressed support for international cooperation in AI governance, saying that an effective regulatory framework requires the involvement of both the United States and China, and added that Germany is pushing for discussions on the issue at the G7 level.