Beacon Policy Advisors LLC — AI Use Policy

1. Purpose

This policy governs the use of artificial intelligence (“AI”) tools by Beacon Policy Advisors LLC (“Beacon”) personnel in connection with Beacon’s work product, client communications, and internal materials. It is intended to allow Beacon to benefit from AI tools where they improve efficiency and quality, while safeguarding the accuracy, integrity, and confidentiality of Beacon’s research and analysis.

2. Permitted Uses

Personnel may use AI tools to:

• Improve the readability, grammar, and organization of drafts (memos, notes, correspondence, internal communications, etc.);
• Condense or tighten written content, provided that all material context, nuance, and substantive meaning are preserved; and
• Assist with routine, non-substantive tasks such as formatting, proofreading, and administrative drafting.

3. Restricted Uses — Substantive Research and Analysis

AI tools may be used to assist with substantive policy research, analysis, and client-facing work product only subject to the following requirements:

• Independent verification of accuracy. Any factual claim, data point, citation, or analytical conclusion generated or suggested by an AI tool must be independently verified by the responsible staff member against a reliable primary or secondary source before it is incorporated into any Beacon work product. AI-generated content may not be relied upon or distributed without such verification.
• No non-public information. Personnel must ensure that AI-assisted output does not incorporate, reproduce, or imply access to material non-public information (“MNPI”), non-public government information, or confidential client information. Staff should independently confirm that any output generated by an AI tool is not, in fact, sourced from or suggestive of non-public information.

4. Prohibited Uses

Personnel may not:

• Input client-confidential information, MNPI, or other non-public or proprietary information into any AI tool that is not approved by Beacon’s chief compliance officer (CCO) for such use;
• Rely on AI-generated analysis, forecasts, or conclusions as a substitute for Beacon’s independent judgment; or
• Represent AI-generated content as original human analysis without appropriate internal review.

5. Approved Tools

Personnel should use only AI tools that have been reviewed and approved by the CCO. Questions about whether a particular tool or use case is permitted should be directed to the CCO.

6. Compliance

Violations of this policy may result in disciplinary action. This policy will be reviewed periodically and updated as AI technology and associated risks evolve.