AI Governance at the AAA

Fair & Ethical AI Design: Design AI systems that align with the AAA’s ethical standards, support fair and unbiased outcomes that respect the rights of all participants, and safeguard equity in dispute resolution. 

Human Centered: Design AI systems with human oversight that are imbued with the AAA’s values, mitigate bias, and preserve control over critical decisions. 

Privacy Enhanced & Data Governed: Manage data responsibly with strict attention to privacy while adhering to data governance standards. 

Secure & Resilient: Foster trust in the dispute resolution process by protecting the AAA’s data confidentiality, integrity, and system reliability while defending against external threats and ensuring operational resilience. 

Accurate & Built with Purpose: AI applications should provide reliable outputs, maintain consistency under varied conditions, and communicate any limitations or the need for human oversight at key points in our processes. 

High Performing: AI applications should adhere to industry standards for uptime and response time to human inputs. 

Explainable AI: The AAA’s AI systems should allow staff and participants in our process to understand and trust their outputs. 

Transparent AI Systems: Clearly communicate the choices we make in developing AI applications to enhance the delivery of services. 

Accountable AI Management: Establish human ownership over AI across the AAA’s processes. 

Vetted & Reliable AI Partnerships: Carefully select and continuously assess third-party vendors to ensure reliable performance and alignment with the AAA’s values. 

Proactive AI Monitoring: Establish and apply standards for monitoring and evaluating AI systems in line with these principles. 

Continuous AI Learning & Improvement: Commit to staff training, research, and feedback loops that maintain our industry-leading culture of innovation and AI literacy.