Trustworthy generative AI covers the safety and privacy aspects during interactions between humans and robots. It focuses on ensuring that AI systems operate transparently, reliably, and ethically to mitigate risks associated with data misuse, biases, and unintended consequences. By prioritizing accountability and explainability, trustworthy AI gains user confidence and aligns robotic behaviors with human values and societal norms.

When it comes to sensitive contexts such as elder care, generative AI must be more careful about their ethical issues to facilitate the elder people. Elderly individuals often exhibit varying levels of comfort with technology and differing expectations regarding privacy and autonomy. Additionally, cultural and individual differences in privacy expectations and ethical standards during interactions with the elders must be considered. Lastly, the reliability of generative AI should be enforced to prevent unexpected outcomes.

In summary, for elder care, trustworthy generative AI plays a crucial role in enhancing the caring quality and reliability for older adults. Achieving trustworthy AI is a rather challenging task to provide companionship and support for elder care. We plan to start from aligning AI on legal compliance to enforce AI comply with existing privacy laws such as the GDPR and HIPAA and safety regulations like the EU AI Act. After legal compliance, we will further work on personalization to let AI meet personal privacy and ethical expectations. Our ultimate goal is to foster trust and acceptance among older adults and their caregivers.