U.S. threats to sanction Chinese AI developers over alleged intellectual property theft and export-control violations risk unravelling efforts to establish a bilateral dialogue on AI safety, analysts say, just as increasingly powerful models raise concerns over serious security breaches.
