The U.S. Chamber’s Technology Engagement Center (“C_TEC”) and Center for Global Regulatory Cooperation (“GRC”) recently released a set of ten principles essential for attaining the full potential of AI technologies.
The principles, drafted with input from more than 50 Chamber member companies, stress the importance of creating a sensible and innovation-forward approach to addressing the challenges and opportunities presented by AI.
The ten principles are:
- Recognize Trustworthy AI is a Partnership: Addressing issues of public concern around AI (e.g., protecting against bias, ensuring democratic values and respecting human rights), must be a multistakholder process involving governments, the private sector, civil society and academics in order to effectively foster public trust in AI technologies.
- Be Mindful of Existing Rules and Regulations: Policymakers should remind themselves that AI technology users must remain accountable under existing laws. In creating new AI policies, governments should adopt a sector-specific approach, avoid creating a patchwork of subnational level AI policies and promote interoperable practices.
- Adopt Risk-Based Approaches to AI Governance: Nations should adopt a flexible, risk-based approach to governing AI technologies rather than prescriptive requirements to ensure appropriate levels of scrutiny based on risk. Industry-led, voluntary accountability mechanisms should focus on addressing harms to individuals that can be empirically linked to the use of AI technologies.
- Support Private and Public Investment in AI Research and Development: To incentivize investment in AI innovation, governments should partner with businesses working at the forefront of AI and promote flexible governance methods, including regulatory sandboxes and utilizing testbeds.
- Build an AI-Ready Workforce: Governments should establish partnerships with different stakeholders, including businesses and educational institutions, to build a workforce ready and capable for an AI economy.
- Promote Open and Accessible Government Data: To accelerate AI development, useful government-owned datasets should be made available and easily accessible while ensuring appropriate privacy and security protections for the data.
- Pursue Robust and Flexible Privacy Regimes: Governments should pursue the creation of robust, yet flexible, data protection regimes that enable the deployment, development and use of AI while also ensuring appropriate privacy protection for consumers. The Chamber refers to its Privacy Principles as a guide for creating such a regime.
- Advance Intellectual Property Frameworks that Protect and Promote Innovation: Although the development of AI often utilizes open source components, governments must ensure full respect for intellectual property rights and appropriately enforce violations. Additionally, companies must not be obliged to transfer or provide access to their source code, algorithms and/or data sets.
- Commit to Cross-Border Data Flows: Governments should refrain from restricting cross-border data flows as much AI adoption is enabled across geographies by cloud computing. Policies such as data localization will act as significant market access barriers and diminish investment and innovation in AI technologies.
- Abide by International Standards: Policymakers should acknowledge and support the development of AI standards in recognized international standards bodies and consortia. Governments should also leverage industry-led standards where possible to facilitate the use and adoption of AI technologies.