The members explored various dimensions essential to formulating an effective AI policy, including:

• Intellectual Property and Data Upload Guidance: The discussion emphasized the importance of clear guidelines on intellectual property rights and best practices for uploading data securely and compliantly.
• Accuracy and Reliability of AI Results: Participants highlighted the need to ensure AI outputs are accurate and reliable, addressing challenges in validating AI-generated results.
• Ownership of Outcomes: It was agreed that users should retain ownership of the outcomes produced by AI tools, reinforcing accountability and transparency.
• Usage of AI Features in Existing Products: The group discussed how AI functionalities integrated within currently used products should be governed, ensuring seamless and ethical adoption.
• Guidance on AI Models to Use: Clear recommendations on selecting appropriate AI models were deemed necessary to align with organizational goals and compliance.
• Ethical and Compliance Considerations: The ethical use of AI, including adherence to compliance standards, was a major theme, with discussions on mitigating risks and ensuring responsible AI deployment.
• Reporting AI Misuse: Establishing robust mechanisms for reporting and addressing AI misuse was identified as a priority to maintain trust and integrity.
• Industry-Specific Policies: The group differentiated between regulated and unregulated industries, discussing tailored AI policies for sectors such as Life Sciences and Consumer Packaged Goods (CPG).
• Cyber Risk and Red Teaming: Participants touched upon cyber risks associated with AI, suggesting this as a topic for a separate, detailed paper. The conversation included risk assessments based on the AI models employed.
• AI Training and Literacy: The importance of ongoing AI education was underscored, with suggestions to implement checklists, annual training, and onboarding programs to enhance AI literacy across the workforce.