AI POLICY

Artificial Intelligence (AI) Use Policy for SEE International

Effective Date: September 11, 2026

Approved By: Executive Leadership

1. PURPOSE

At SEE International (hereon referred to as “SEE”) we are committed to leveraging the power of Artificial Intelligence (AI) to maximize our impact and enhance efficiency, and do better work toward our mission, while respecting the rights and dignity of all individuals ensuring that decisions are made with consideration for their potential impacts on society. We will prioritize responsible use, data security, and ethical considerations as we embrace AI technologies. This policy outlines the guidelines around using generative AI tools in compliance with our handbook and as an extension of our company values.

2. SCOPE 

This policy applies to all: employees, contractors and consultants, volunteers, and board members. This policy also is in effect when using AI tools on behalf of SEE, whether using organization-provided or personal accounts.

3. DEFINITION OF ARTIFICIAL INTELLIGENCE (AI)

Artificial Intelligence refers to software or systems that generate content, analyze data, automate tasks, or assist decision-making using machine learning, natural language processing, or similar technologies.
Examples include (but are not limited to):

  • Generative text tools (e.g., drafting emails, grant language)
  • Image or design tools
  • Data analysis or forecasting tools
  • Chatbots or automated responses
4. GUIDING PRINCIPLES

AI use at SEE must adhere to the following principles:

  • MISSION ALIGNMENT: AI use must support—not replace—our mission and human judgment.
  • TRANSPARENCY: Stakeholders should not be misled about the use of AI.
  • PRIVACY / CONFIDENTIALITY: Sensitive, personal, donor, or proprietary information must be protected.
  • EQUITY / FAIRNESS: AI tools must not reinforce bias, discrimination, or inequity.
  • HUMAN OVERSIGHT: AI outputs must be reviewed by a human before use.
5. ACCEPTABLE USES

All managers will implement rules for acceptable use, and all employees must demonstrate proper use. AI tools may be used for:

  • Drafting or editing non-confidential content (e.g., newsletters, marketing copy, first draft narratives)
  • Brainstorming ideas, outlines, or strategies
  • Summarizing public information or internal documents (without sensitive data)
  • Improving grammar, clarity, or tone
  • Supporting research and trend analysis (with independent verification)
  • All AI-generated content must be reviewed, edited, and approved by staff before publication or use.
6. PROHIBITED USES

AI tools will not be used for:

  • Uploading or processing confidential information, including:
    • Donor data and personally identifiable information (PII) will not be shared with generative AI software, except for the narrow, defined email engagement data exception described in the Authorized Engagement Data Integration section below.
    • Making final decisions related to:
      • Hiring, firing, or performance evaluations
      • Donor eligibility or benefits
      • Program eligibility determinations
      • Creating deceptive content (e.g., impersonation, undisclosed AI-generated communications)
      • Replacing required professional judgment, legal advice, or fiduciary responsibility
7. DATA PRIVACY & SECURITY

Staff must not input confidential or sensitive data into public AI tools unless explicitly approved. AI tools must comply with applicable data protection laws and organizational privacy policies. Any suspected data breach or misuse involving AI must be reported immediately to SEE’s CEO.

8. AI PILOT & SANDBOX PROGRAMS

To encourage responsible innovation while maintaining security, SEE allows staff to propose limited AI Pilot Programs to test new AI workflows, tools, or platform integrations before full adoption.

  • Approval Requirement: All AI pilots, including third-party integrations, new software subscriptions, or novel data workflows, must receive prior written approval from the Systems and Performance Manager and relevant department leadership before enablement.
  • Pilot Duration & Scope: Pilots are approved for a single user or defined project team and a limited timeframe (typically 90 days). Unrestricted or organization-wide rollouts are prohibited during the pilot phase.
  • Data Protection Baseline:
    • Donor data and PII remain off-limits unless anonymized (e.g., using system Contact IDs).
    • AI vendor model training must be explicitly disabled in account settings.
    • System access must be set to read-only whenever technically feasible.
    • Logging & Retention: Staff running a pilot must maintain a log of use cases and delete temporary project data, exports, and chat histories within 90 days of pilot completion.
    • Review & Transition: At the conclusion of the pilot period, the project will be evaluated. To transition a pilot into permanent policy or expand it organization-wide, a formal policy review/amendment is required under Section 13.
9. AUTHORIZED ENGAGEMENT DATA INTEGRATION

SEE has approved specific, limited connections between Claude and our marketing engagement platforms, to support analysis and reporting on campaign performance and to generate contact lists for approved campaign purposes under pilot program safety controls. However, donor names, email addresses, mailing addresses, payment or financial account information, and all other donor or constituent data beyond a bare numerical identifier and engagement data remain off-limits, consistent with Section 6. This is a narrow, defined exception to the general prohibition on sharing donor or PII data with AI tools (Section 6, Prohibited Uses) — it is not blanket approval to connect Claude to any other system or dataset. Two categories of use are authorized: (1) aggregate, segment-level engagement analysis, and (2) per-contact list generation using marketing engagement platform numerical identifier in place of names or other direct identifiers, subject to required controls during the AI Pilot period.

10. DISCLOSURE AND ATTRIBUTION

When appropriate, SEE must disclose the use of AI in content creation, especially when transparency is important to donor trust, public communications, or educational materials. AI should be viewed as an assistive tool, not an author or decision-maker.

11. TRAINING AND OVERSIGHT

SEE will provide guidance or training on responsible AI use as needed. Supervisors are responsible for ensuring AI use aligns with this policy. This policy will be reviewed periodically as AI technology and regulations evolve.

12. MONITORING AND EVALUATION

We will continuously monitor the performance of our AI systems, assessing their impact on our operations and the communities we serve. Regular evaluations will inform improvements and adjustments to our AI strategies.

13. VIOLATIONS

Failure to comply with this policy may result in corrective action, up to and including termination of employment or volunteer engagement, depending on severity.

14. POLICY REVIEW

This policy will be reviewed at least annually or as legal, ethical, or technological changes require.

15. CONTACT AND REPORTING

For inquiries or to report a concern related to our AI use, please contact communications@seeintl.org. We encourage open communication to ensure responsible use of AI in our organization.

Get Our Monthly Updates

This field is for validation purposes and should be left unchanged.