Secure architectures for AI agents - How autonomous LLM agents gain reliable control

Great technologies, programming languages and tools are used to create software. That is good and right. But unfortunately we often lose sight of the fact that the decisive factor is not the technology, but the domain.