Responsible Use of AI

The AI regulatory landscape is rapidly developing, and there are several challenges and barriers to overcome before the full potential of digital health can be achieved. We support the development of AI laws and regulations, and we recognize there is a need for a multitude of stakeholders to be part of the AI legislative and rulemaking processes. We believe AI requirements must ensure that constituents use AI in a responsible manner, particularly for biopharmaceutical manufacturers, to bring the best medicines to patients in a safe and effective way.

All development, deployment, and use of AI at Lilly is guided by the following principles:

• Accountable – We hold ourselves accountable through robust governance, clear lines of responsibility, monitoring, and continuous improvement. We expect our third parties, vendors, and partners to do the same.

• Safe – We seek to protect against unintended applications, unintended access, and harm associated with the use of AI.

• Valid and Reliable – We aim to employ technology that produces appropriately consistent and accurate outputs compared to standard human processes and includes controls that minimize unforeseen disruptions.

• Fair and Equitable – We strive to use AI in a way that mitigates any inherent bias and discrimination in the data, information, or algorithms.

• Privacy-Enabled and Protected – We strive to safeguard personal information, confidentiality, autonomy, and control over the use and disclosure of privacy information and intellectual property.

• Transparent and Explainable – We promote transparency regarding the design, function, and use of AI and endeavor to explain, show, or describe why and how results have been produced when feasible and appropriate.

• Sustainable – In cases where we have control over the production of AI, we are committed to reducing our environmental footprint and promoting the conscientious use of AI.

• Enhanced – We will optimize our scientific and medical expertise to promote safe and effective AI in healthcare.

AI Governance

Executive Oversight and Accountability

Our AI governance framework is anchored by executive leadership and clear lines of accountability. The Executive Vice President and Chief Information and Digital Officer serves as the owner of our Using AI Responsibly procedure that is applicable to all employees and contractors. This procedure establishes the basic principles mentioned above, requires employees to use AI appropriately, and empowers a cross-functional team to intake new AI use cases for review against our requirements, industry standards, and legal obligations.

This cross-functional team also supports Lilly's use of AI by working directly with teams across the company on how to appropriately use AI, creating guidance for specific AI use cases, driving risk impact assessments, and directing teams to existing enterprise AI capabilities where appropriate.

Our Board of Directors, through the Audit Committee, provides oversight of the company’s programs, policies, procedures, and risk management activities related to information security, cybersecurity, and data protection, which includes AI-related risks. The Audit Committee receives periodic briefings on responsible use of AI at Lilly.

AI Use Case Review Process

Before deploying new AI solutions or applying new use cases for existing AI at Lilly, employees must follow a structured review process. This includes reviewing Lilly's approved and standard AI platforms, obtaining advanced approval for new AI solutions or use cases, closely reviewing inputs, algorithms, and outputs for bias and errors, and ensuring compliance with applicable regulations and Lilly policies. The AI use case review process is in addition to and does not replace other required processes for development and deployment of new technologies, including privacy reviews, information security reviews, and third-party procedures.

Third-Party and Vendor Management

We hold our third parties, vendors, and partners to the same responsible AI standards we set for ourselves. Any agreements with vendors who use AI require specific clauses to ensure AI quality, define clear roles, and protect our rights and data. Third parties who create AI retain the right to update their acceptable use policies, and employees must comply with applicable vendor policies when using third-party AI tools.

Human Oversight and Employee Accountability

AI tools support work at Lilly but do not replace human creativity, expertise, autonomy, agency, or oversight. Employees are always responsible for the accuracy and appropriateness of all information generated using an AI system. All AI outputs must be verified for accuracy, and any content intended for external dissemination must be reviewed and approved following appropriate procedures. We prohibit the use of AI to replace human autonomy or to generate content without necessary permissions and consent.

Risk Management and Monitoring

Our approach to AI risk management is integrated with our broader enterprise risk management framework. We conduct risk impact assessments for AI implementations and require bias audits where appropriate, particularly in clinical, HR, promotion, or other high-risk settings. Proper testing, implementation, and ongoing monitoring of AI systems ensure proper deployment and help identify issues early.

Policy Positions

We maintain public policy positions on topics where AI intersects with our business, including data privacy and security, AI in drug discovery and development, AI in clinical trials, and AI and intellectual property. These positions reflect our view that regulatory frameworks should promote responsible innovation while maintaining the safety and efficacy standards patients depend on. Our full AI policy positions are available here.

In 2024, we established the foundational framework for responsible AI use at Lilly with the introduction of a global procedure that governs how AI is developed, deployed, and used across the organization. That framework sets clear expectations for employee accountability, disclosure practices, and compliance with vendor policies. Our AI Team provides ongoing guidance, conducts risk impact assessments, and maintains a curated set of approved AI products available to employees. We continue to evaluate and approve AI tools for specific use cases while maintaining rigorous data protection standards and integrating AI governance with our broader ethics and compliance program.

As AI technologies continue to evolve, we are focused on keeping our governance practices current. We routinely refine our internal procedures as new technologies, regulations, and industry standards emerge. We also participate in broader industry and regulatory discussions on AI governance, and we support the development of federal frameworks that promote innovation while maintaining the safety and efficacy standards patients depend on.

Resources

• Lilly AI Policy and Implications for Digital Health and Drug Development

• Lilly Position Statement on Artificial Intelligence (AI)

Related Topics

• Ethics and Compliance

• Corporate Governance

• Innovate

• Data Privacy

• Bioethics

• Patient Safety

• Employee Learning and Development

• Supply