The UNCG Artificial Intelligence Oversight Committee urges the Spartan community to adopt the following Responsible AI Principles to guide university procurement, development, implementation, and monitoring of AI-enabled tools within its provision of academic and business services:
Accuracy and Dependability
AI-enabled tools should be developed using high-quality, representative data to ensure outputs are accurate, fair, and reliable. Ongoing testing, monitoring, and continuous improvements should be in place to maintain dependable performance as AI systems evolve with new data and dynamic conditions.
Privacy and Security
AI-enabled tools should be designed to minimize potential harm to users, society, and the environment, with transparent and understandable decision-making processes to build trust and protect the privacy and security of individuals and their personal data. Proactive failure management strategies should be in place to mitigate risks, alert stakeholders, and ensure a quick return to safe operations.
Accountability
AI-enabled tools should have well-defined accountability structures, with clear responsibility for their development, deployment, and outcomes, and should have independent ethical oversight to guard against misuse and ensure responsibility for any adverse consequences.