GS-23-001 Artificial Intelligence Responsible Use Guidelines

Effective Date: 12/12/23

PURPOSE

GTA intends to provide a roadmap that guides state agencies in maximizing the benefits of these technologies while minimizing risks and ensuring that AI remains a force for good in the service of Georgians. These guidelines outline a comprehensive framework for the responsible use of AI and generative AI systems within state agencies.  By adopting these principles, state agencies can maintain public trust, drive innovation, and contribute to a more equitable, efficient, and forward-thinking public sector.

SCOPE and AUTHORITY

O.C.G.A 50-25-4(a)(8) – State Government, Georgia Technology, General Powers

O.C.G.A 50-25-4(a)(20) - State Government, Georgia Technology, General Powers

PM-04-001 – Information Technology Policies, Standards and Guidelines

PS-08-005 – Enterprise Information Security Charter

GUIDELINES

Principles

In order to safeguard the welfare and enhance the services provided to the citizens of Georgia, the GTA has established five guiding principles governing the design, implementation, and utilization of automated systems. Informed by industry research and experts, these principles are intended to direct state agencies aiming to integrate protective measures into their policies and operational procedures. These principles serve as a framework whenever automated systems have significant implications on opportunities, the rights of Georgia residents, or their access to essential services.

I. Implement Responsible Systems

1. User-Centered Design and Development

State agencies should prioritize user research as an integral component in the procurement or development of automated systems. It is important to maintain the human element during the design of any service. Seek input and insights from user groups, diverse stakeholders, and domain experts to identify concerns, risks, and potential impacts associated with the system.

2. Comprehensive Testing and Monitoring

Automated systems must undergo pre-deployment user testing to identify potential risks and assess their intended functionality. Implement risk identification and mitigation strategies to ensure system safety and effectiveness, including addressing unintended consequences.

3. Ongoing Monitoring: Ongoing monitoring is essential to confirm that the system operates as intended, and deviations should be addressed promptly. Adhere to domain-specific standards to ensure compliance and compatibility with industry best practices.

4. Consideration for Non-Deployment: State agencies should be prepared to halt the deployment of an automated system or remove it from use if it fails to meet safety or effectiveness standards.

5. Data Protection: Ensure that the design, development, and deployment of automated systems protect against inappropriate or irrelevant data use. Mitigate the risks associated with the reuse of data, preventing compounded harm.

6. Independent Evaluation and Transparency: The GTA will reserve the right to conduct an independent evaluation and reporting to confirm the safety and effectiveness of automated systems, including efforts taken to mitigate potential harm. The GTA will make evaluation results publicly available whenever appropriate, promoting transparency and accountability.

II. Ensure Ethical and Fair Use of Automated Decisions

1. Prioritize fairness, transparency, accountability, and privacy: State agencies should adopt a set of ethical AI principles that prioritize fairness, transparency, accountability, and privacy in the design and deployment of AI systems for state services. Develop a set of ethical guidelines for AI system design and deployment. They serve as a compass to ensure ethical practices throughout the AI lifecycle.

2. Algorithmic Bias Awareness: Raise awareness among agency staff about the concept of algorithmic bias and its potential impacts on decision-making processes within state services. Provide training and educational programs for agency staff to ensure they understand what algorithmic bias is and how it can impact decision-making. Awareness is the first step in addressing bias effectively.

3. Data Quality and Diversity: Carefully curate and vet the data used to train AI algorithms. Make sure the data is diverse and representative of all relevant demographic groups. This helps prevent biased outcomes caused by skewed data. Ensure that the data used to train AI algorithms is representative and diverse, preventing underrepresentation or bias against any demographic group.

4. Ongoing Monitoring: Continuously monitor AI systems to detect and rectify biases as they emerge. Regular assessments are essential to maintaining fairness and effectiveness over time. Implement continuous monitoring mechanisms to assess AI systems for potential biases and discriminatory outcomes during their operation.

5. Bias Mitigation and Impact Assessment: Develop and implement strategies to address and mitigate algorithmic bias whenever detected, such as refining algorithms, adjusting data inputs, or retraining models. Regularly assess how AI systems affect different user groups. Understand any disparities or unintended consequences that may arise and take action to rectify them.

6. Transparency and Accountability: Make sure that the decision-making processes within AI systems are transparent and explainable. Users should be able to understand why a particular decision was made, instilling trust in the system. Assign specific individuals or teams responsible for monitoring AI systems for bias and taking corrective actions. Clearly defined accountability ensures that bias-related issues are addressed promptly.

III. Maintain Data Quality and Privacy

1. Data Governance Framework: Establish clear guidelines for data governance to maintain integrity and privacy.

2. Security and Transparency: Prioritize robust security measures and transparent data handling practices.

3. Data Quality and Retention:  Ensure data accuracy, minimize storage, and dispose of obsolete data.

4. Compliance and Accountability: Maintain compliance with data protection laws, conduct regular audits, and involve the public in the decision-making process.

IV. Keep AI Usage Transparent

1. Transparency in System Use: Ensure that individuals are informed about the use of automated systems and understand how these systems contribute to outcomes that affect them.

2. Accessible Documentation: Encourage designers, developers, and deployers of automated systems to provide plain language documentation that is easily accessible to the public. This documentation should include clear descriptions of system functionality, the role of automation, system ownership, and explanations of outcomes.

3. Up-to-Date Notices: Require that notice regarding the use of automated systems is kept current, and individuals impacted by the system should be notified of significant use case or key functionality changes.

4. Explanation of Outcomes: Ensure that individuals have access to information explaining how and why outcomes that affect them were determined by automated systems, even when these systems are not the sole contributors to the outcome.

5. Technically Valid Explanations: Mandate that automated systems provide technically valid, meaningful, and useful explanations to affected individuals, as well as operators and stakeholders who need to understand the system. The level of detail in these explanations should be contextually calibrated to the level of risk involved.

6. Public Reporting: Promote the publication of summary information about automated systems in plain language. Assessments of the clarity and quality of notice and explanations should also be made public whenever possible to enhance transparency and public trust.

V. Keep Human Involvement at the Center

1. Human Responsibility and Ownership in AI Systems: State agencies shall establish and adhere to policies that emphasize human responsibility and ownership of the outcomes produced by AI systems used in state services. This policy outlines the key principles to guide state agencies in achieving these objectives.

2. Transparency and Accountability: State agencies shall prioritize transparency and accountability in the deployment of AI systems. Agencies must maintain clear records of AI system use, their objectives, and the roles of individuals overseeing and interacting with these systems.

3. Human Oversight and Decision-Making: AI systems should not operate in isolation. State agencies must ensure that humans retain control over the operation of AI systems and that human decision-makers remain responsible for the final decisions made with the support of AI technology.

4. Ethical Design and Use: Agencies shall mandate that AI systems be designed and used in accordance with ethical principles that prioritize fairness, transparency, accountability, and privacy. Ethical considerations must be an integral part of AI system development and usage.

5. Clear Roles and Responsibilities: Clearly define roles and responsibilities for individuals involved in AI system implementation. This includes specifying the duties of AI system operators, data stewards, and decision-makers.

6. Human-AI Collaboration: Encourage collaboration between humans and AI systems to enhance decision-making processes. AI should be viewed as a tool that complements human expertise rather than a replacement for human judgment.

7. User Education: Promote user education to ensure that individuals interacting with AI systems understand the capabilities and limitations of these technologies. Users should be aware of how AI contributes to outcomes and that humans remain responsible for those outcomes.

8. Training and Competency: Agencies must invest in training and development programs to equip their staff with the skills and knowledge necessary to effectively use AI systems and make informed decisions.

9. Ownership of Data and Models: State agencies shall retain ownership and control over the data used to train AI models and the models themselves. This ownership ensures that AI systems serve the agency's mission and values.

10. Continuous Monitoring and Improvement: Establish mechanisms for continuous monitoring and assessment of AI systems. Regularly evaluate system performance, ethical adherence, and the impact on outcomes, and take corrective actions as needed.

RELATED ENTERPRISE POLICIES, STANDARDS AND GUIDELINES

Artificial Intelligence (AI) Responsible Policy PS-23-001

Artificial Intelligence Responsible Use Standard SS-23-002

 

 

 

Related Files