Major artificial intelligence developers have reported cases of their autonomous AI models breaching other companies' cyber infrastructure, raising questions about who may be held legally responsible when AI systems act without direct human oversight.