O.C. Tanner AI Policy & Ethics Principles (2024)

Introduction

Artificial intelligence (“AI”) technologies come in various forms and can enhance workplace efficiency while reducing costs. O.C. Tanner recognizes the potential of AI technologies, but their deployment must be responsible, lawful, and consistent with O.C. Tanner’s policies.

O.C. Tanner has adopted a Policy that governs O.C. Tanner’s use of products, services, tools, and technologies that incorporate or leverage AI, machine learning, and/or generative AI (collectively referred to as “AI Tools”). The Policy is further described below.

Purpose

The purpose the Policy is to set out the rules for the responsible use of AI Tools while protecting O.C. Tanner and its clients and mitigating the risks of misuse, unethical outcomes, potential biases, inaccuracies, and unauthorized disclosures of personal, confidential, sensitive, or proprietary information.

The Policy ensures that the use of AI Tools in the workplace is authorized, monitored, ethical, and compliant with contractual obligations and applicable law, as well as other company policies and practices.

Scope of Policy

The Policy applies to all O.C. Tanner employees who interact with, develop, or implement AI Tools within our organization. It covers use of AI Tools related to products and services O.C. Tanner provides to clients, as well as interactions with clients, third parties, and suppliers.

Compliance with related policies and agreements

The Policy is intended to add to, not contradict, limit, or replace, applicable mandatory rules, policies, legal requirements, and contractual obligations, all of which remain in full force and effect. Any use of AI Tools under the Policy must comply with other relevant O.C. Tanner policies.

Rules for using AI Tools

O.C. Tanner has established an AI Committee that is responsible for reviewing and approving AI Tools (in the workplace and within our software services). All potential uses involving AI Tools must be approved by the AI Committee.

Without written authorization from the AI Committee, O.C. Tanner will not:

• enter any O.C. Tanner, employee, client, or third-party confidential, trade secret, or other personal, confidential, sensitive, or proprietary information into a prompt for an AI Tool.
• create work product for clients using an AI Tool.
• use a personal or publicly available version of an AI Tool (e.g., a non-enterprise version) for any work-related functions.

Where use of an AI Tool is authorized by the AI Committee, O.C. Tanner employees will:

• exercise good judgment when using the AI Tool. Output from an AI Tool is best used to complement other information, knowledge, or tools, with human oversight to provide consistency and control.
• thoroughly review all output from an AI Tool before using it or forwarding it to other O.C. Tanner employees to verify: (i) that outputs do not contain biased, offensive, or discriminatory content; (ii) that outputs do not improperly use or disclose personal or confidential information; and (iii) the accuracy or reported facts with other trusted sources.
• comply with any product-specific rules based on legal review of the relevant terms and conditions governing O.C. Tanner’s use of the AI Tool.
• use the enterprise version or license that O.C. Tanner has secured for the employee.
• use the AI Tool in compliance with the Policy, including the AI Ethics Principles.

The Policy prohibits O.C. Tanner employees from: (i) entering content into an AI Tool that is offensive, discriminatory, or otherwise inappropriate; or (ii) using AI Tools to conduct or solicit illegal activities, infringe on the rights of others, including privacy and intellectual property rights, or interfering with the performance of their jobs or the jobs of others.

The O.C. Tanner AI Ethics Principles (“Principles”) set out below provide a foundation for the responsible development, deployment, and use of AI Tools in the context of (i) deploying AI Tools in the services or software that O.C. Tanner provides to its clients; and (ii) using AI Tools to perform employment related tasks or functions. These Principles guide O.C. Tanner’s and its employees’ use of AI Tools while minimizing potential harms and maximizing the positive impacts for our employees and clients.

1. Privacy by Design. Privacy by design means integrating privacy into products and services by default. When deploying an AI Tool in products or services, privacy by design must be considered through all stages of development. AI Tools must not be deployed in products or services offered to O.C. Tanner clients unless the client has been informed of the AI Tool.
2. Transparency. Ensure transparency in the design, development, and deployment of an AI Tool. Clearly communicate the capabilities, limitations, and potential risks associated with the AI Tool to stakeholders and affected clients.
3. Data Protection. Respect privacy rights and protect personal data when using an AI Tool. Robust data protection measures and contractual terms must be in place with third party providers of AI Tools. Minimize the collection of personal data to the extent possible.
4. Fairness and Equality. Promote fairness and prevent bias in AI Tools. Ensure that output from an AI Tool does not discriminate or disadvantage individuals or groups based on characteristics such as race, gender, religion, or socioeconomic status.
5. Social Benefit. Develop and/or deploy AI Tools for the benefit of helping people thrive at work. Align AI Tools with broader human resources and employee recognition goals, such as, reducing inequality and bias in the workplace.
6. Human Control and Involvement. Ensure that O.C. Tanner employees retain ultimate control over AI Tools deployed in products or services. Avoid the development or deployment of an AI Tool that results in meaningful decision-making without human involvement. When used to perform employment related functions or tasks, carefully and thoughtfully review output. Raw output from an AI Tool must never be presented or delivered to others as finished work product.
7. Accountability. Hold developers, operators, and organizations responsible for the outcomes and impacts of AI Tools.
8. Reliability. Prioritize the reliability of AI Tools. Develop and/or deploy AI Tools with rigorous testing, validation, and risk mitigation procedures to minimize potential harm to O.C. Tanner, its employees, and its clients. Implement contingency plans to address system failures.
9. Ethical Use. Develop and/or deploy AI Tools in an ethical manner consistent with legal requirements. Do not develop or deploy AI Tools for malicious purposes, including manipulation or harm to O.C. Tanner, its employees, and its clients.
10. Monitoring and Assessment. Regularly monitor and assess the impacts of AI Tools used by O.C. Tanner employees and/or deployed in products or services. Conduct ongoing evaluations to identify potential biases, unintended consequences, or risks, and take appropriate actions to address them. Collaborate with relevant stakeholders to continually address the challenges and implications of AI Tools.