Germany Rejects Halting AI Development as Europe Debates Safety

Germany has said that halting the development of artificial intelligence is not a realistic option for Europe, as governments across the region weighed calls from a leading tech executive to slow the pace of AI work and focus more on testing.

The debate over AI safety drew fresh attention after Anthropic CEO Dario Amodei said over the weekend that frontier AI companies should face independent scrutiny and work toward common standards as increasingly powerful systems emerge.

Germany's digital affairs ministry said stopping development was not viable. A spokesperson said strengthening Europe's digital sovereignty would require continued AI development and innovation. Berlin is monitoring global AI developments closely and takes public warnings from leading developers seriously, the spokesperson added.

The ministry also pointed to a new kind of risk: AI systems capable of breaking out of test environments and accessing other systems on their own, as seen in July when OpenAI agents hacked the open-source platform Hugging Face. Such scenarios would represent an entirely new threat requiring a new policy response, the spokesperson said.

Germany backed international coordination on AI governance, saying effective oversight would need the participation of both the United States and China. Berlin is raising the issue at the G7 level.

In Spain, Digital Transformation Minister Oscar Lopez said discussion was growing about the need for an international agreement to manage AI risks, comparing some proposals to nuclear non-proliferation arrangements. He said AI offered enormous opportunities, including discovering new medicines and improving treatments, but also posed significant risks in areas such as finance and cybersecurity. Lopez criticised U.S. President Donald Trump for dismissing concerns raised by AI researchers and developers.

The European Commission said innovation and safety must go hand in hand. A Commission spokesperson said the EU supports developing AI innovation, but companies must prove their services are safe for citizens before operating in the bloc.

Britain welcomed the debate among leading AI companies over the risks of increasingly powerful systems, but said any future regulatory response should be driven by evidence rather than speculation. A spokesperson for British Prime Minister Andy Burnham said it was good that companies developing the most advanced AI systems were now talking openly about the risks as well as the opportunities, while cautioning that the existing framework should not be assumed to remain sufficient as AI capabilities develop.

Britain, Europe's largest hub for AI investment and startups, has generally favoured a lighter-touch approach to regulation, aligning more closely with the United States than with the European Union.