Germany taking AI warnings from US tech leaders 'very seriously'

Germany's National Security Council has twice looked into the threat posed by artificial intelligence getting out of control, a government spokesman said on Monday, following warnings from several US tech executives.

"We take the warnings we have heard very seriously," deputy government spokesman Steffen Meyer told journalists. He added that Berlin was calling for global solutions.

Meyer highlighted both the major opportunities presented by AI, in areas such as the economy, productivity and medicine, as well as the associated risks and dangers. 

"That is why it is important that we do both: We should not focus only on doomsday scenarios, but instead seize the opportunities while also acting decisively wherever we identify risks," he said.

He called for solutions to be pursued within the framework of the G7 group of leading economies and elsewhere.

A Digital Affairs Ministry spokesman described the AI threat as a "paradigm shift" to which the government was responding.

"If AI models become increasingly powerful, break out of testing environments and independently gain access to other systems, we are dealing with an entirely new threat scenario," he said.

The spokesman pointed to a newly established National AI Safety Institute, which is intended to identify security-related risks associated with AI at an early stage.

Over the weekend, Anthropic chief executive Dario Amodei renewed calls to slow down AI development. He was supported by rivals in the sector, including Elon Musk and Sam Altman of OpenAI, the developer of ChatGPT.