To fully unlock AI’s potential, organizations should move beyond fear or complacency and embrace a framework that balances innovation with responsibility. Achieving this balance requires robust AI governance, seamless integration of risk management into existing processes, and a culture that values transparency, fairness and accountability. In this guest article, Pari Sarnot, who, at the time of writing, was a member of the cyber privacy and risk advisory practice at Grant Thornton Advisors LLC, and is currently a product privacy manager at Meta, offers a practical roadmap of key considerations for establishing an AI governance program, including the role of trustworthy AI and steps to take in the AI solution lifecycle development process. The views Sarnot expresses in this article are her own and do not represent those of current or previous employers. See “DOJ’s 2024 Edits to the ECCP: Some History and AI Expectations” (Nov. 6, 2024).