Last updated 09/24/26
At Flynn, one of our core beliefs has long been “continuous improvement through learning.” That commitment encompasses many forms of education, experimentation, and growth, including the responsible adoption of emerging technologies such as artificial intelligence (AI).
Flynn encourages employees to explore how AI can improve efficiency, effectiveness, creativity, and the quality of our work. As with any technology, however, clear guidelines are necessary to ensure AI is used responsibly, securely, ethically, and in a manner that protects Flynn and our clients.
This policy applies to all employees, contractors, freelancers, interns, and others performing work on behalf of Flynn.
Approved AI Tools
Flynn’s internal AI team evaluates AI technologies and determines which tools are appropriate for agency use. Approved tools, access methods, and recommendations may change as technology and associated risks evolve.
Current approved tools include, but are not limited to:
- Specific AI tools: Figma, Anyword, Midjourney, ChatGPT, Claude
- AI functionality incorporated into approved platforms: Microsoft Copilot, Adobe Creative Cloud, and other approved agency software
Employees may only use Flynn-approved AI tools and Flynn-provided accounts for client work. Personal AI accounts may not be used for Flynn or client work.
Any employee interested in using an AI tool that has not been approved must obtain approval from Flynn’s AI team before downloading, connecting, purchasing, or using the tool for agency business.
Responsible AI Usage
AI is a tool to assist and enhance our work—not a substitute for human thinking, judgment, strategy, or creative originality. Final work produced by Flynn remains the responsibility of Flynn employees, regardless of whether AI contributed to its development.
All AI-assisted work must be reviewed, curated, and refined by a Flynn employee before being shared internally, presented to a client, published, or distributed.
To protect the quality and integrity of our work:
- Accuracy: AI-generated information, claims, statistics, citations, sources, and other factual content must be independently verified before use.
- Quality and brand alignment: AI-assisted work must meet Flynn’s quality standards and appropriately reflect Flynn’s or the client’s intended voice, tone, strategy, and brand standards.
- Bias and appropriateness: AI outputs must be reviewed for bias, discrimination, inclusivity, appropriateness, and potentially misleading, defamatory, or offensive content.
- Human judgment: AI may assist with research, brainstorming, analysis, drafting, automation, and concept development, but employees remain responsible for the strategic and creative decisions behind the work.
- Intellectual property: Employees may not prompt AI tools to imitate or replicate the distinctive work or style of a specific living artist, creator, photographer, illustrator, writer, or other creative professional.
- Copyright: Purely AI-generated material may not qualify for copyright protection under current U.S. copyright law. AI-generated material intended for use in a final client deliverable must therefore include sufficient human authorship, creative decision-making, modification, and/or original proprietary input appropriate to the work and its intended use.
- Review and accountability: Use of AI does not change Flynn’s standards for accuracy, quality, legal compliance, client approval, or employee accountability.
Client Permissions and Disclosure
Flynn respects that clients may have different policies, contractual requirements, risk tolerances, or preferences regarding the use of AI. Client-specific requirements always take precedence over Flynn’s general permission to use an approved AI tool.
Clients may request that AI not be used on their account or for particular projects. Such requests must be documented and honored.
When AI is materially involved in developing concepts or executions presented to a client, the use of AI should be disclosed when appropriate to the nature of the work, the client relationship, contractual requirements, or applicable law or platform policy.
AI-Generated Content, Advertising and Social Media
AI-generated and materially AI-altered content used in advertising, social media, or other public-facing communications must comply with all applicable laws, regulations, client requirements, and platform-specific disclosure policies.
Employees responsible for creating, approving, trafficking, or publishing such content must determine whether an AI disclosure or platform designation is required before publication.
This includes, but is not limited to:
- Realistic AI-generated or materially altered images or video
- Synthetic or cloned voices
- AI-generated music or audio
- Content that depicts a real person saying or doing something that did not occur
- AI-generated depictions of real events, locations, products, services, or circumstances that could reasonably be mistaken for authentic footage or photography
- AI-generated people, testimonials, endorsements, demonstrations, or representations of customer experiences
- Material use of a person’s name, image, voice, or likeness
Where a platform provides an AI-generated, altered-content, paid-partnership, branded-content, or similar disclosure mechanism, employees must use that mechanism whenever required by the platform, law, regulation, or client policy.
AI must never be used to create a false testimonial, fabricated endorsement, misleading product demonstration, fictitious consumer experience presented as real, or deceptive representation of a product or service.
The use of AI does not eliminate or replace any other advertising disclosure requirements, including disclosures relating to sponsorships, endorsements, influencers, material connections, regulated products or services, or other legally required advertising disclosures.
AI Note-Taking, Transcription and Meeting Tools
AI-powered recording, transcription, note-taking, or meeting-assistant tools may only be used for client communications when appropriate permission has been obtained.
When using these tools:
- Participants must be informed when a meeting is being recorded or transcribed and consent must be obtained when required by applicable law or client policy.
- Recording or transcription must be stopped if a client or participant requests it.
- Recordings, transcripts, and notes must be stored only in approved systems and made accessible only to authorized individuals.
Data Privacy and Security
Confidential, sensitive, proprietary, or personally identifiable information must not be entered into an AI system unless the system has been specifically approved by Flynn for that type of information and its use is consistent with applicable client agreements, privacy requirements, and Flynn security policies.
Unless specifically authorized, employees must not enter or upload:
- Financial information: Credit card numbers, banking information, non-public financial statements, or other sensitive financial data
- Personally identifiable information: Social Security numbers, home addresses, personal phone numbers, personal email addresses, government identification numbers, or other sensitive personal information
- Confidential business information: Trade secrets, confidential strategies, customer lists, non-public business information, proprietary code, API keys, passwords, authentication tokens, or private URLs
- Health information: Protected health information, medical records, diagnoses, medications, treatment information, or other confidential health data
Employees should minimize the amount of client-identifying information included in AI prompts and inputs whenever possible.
Compliance and Policy Updates
AI technologies, laws, regulations, and platform policies are evolving rapidly. Flynn will periodically update this policy and its list of approved tools to reflect changes in technology, industry standards, legal requirements, client expectations, and best practices.
Employees are responsible for following the most current version of this policy and any additional guidance issued by Flynn.
Violations of this policy—including misuse of confidential information, infringement of intellectual property rights, failure to comply with disclosure requirements, or other inappropriate use of AI—may result in revocation of AI tool access and corrective or disciplinary action as determined by further HR review.