While developers have a critical role in creating AI systems, the primary responsibility for AI-caused harm should lie with the deployers. They are the ones choosing to implement AI in specific contexts and must ensure it is used safely and ethically. Developers cannot foresee every use case or potential misuse.