PR Agency One provides guidelines for using generative AI tools in public relations campaigns. This information was correct as of 13 July 2023 but may need updating as new technology emerges and thinking changes.
We promise to:
- Protect the confidentiality of client information by refraining from using generative AI tools to create confidential content. This includes client business plans, PPTs or documents, paid analyst reports, market insights, confidential research data, and text related to sensitive internal employee communications.
- Use proxy and anonymous company and personal details to maintain privacy. Additionally, avoid to use generative AI images as final creative for a client campaign as it may be at risk of copyright infringement and may limit the user’s ownership rights in the work generated. These guidelines also apply to confidential internal agency information.
- Declare to clients all AI tools which are used by the company, should they wish to know
- Avoid using generative AI to create or spread deepfakes, misinformation, or disinformation, and honour the role of PR agencies in society.
- Commit to accuracy by checking and sourcing data generative AI tools provide, validating claims with a search of the source, checking for inadvertent plagiarism, copyright infringement, or trademark infringement in AI-generated output.
- Ask vendors about how they use AI in their tools and work to eliminate biases and improve accuracy. Also, ask vendors to provide transparency around their prompts and inputs used to generate desired output.
- Ensure transparency by disclosing the use of generative AI tools in any part of the creative, reporting and decision making process to clients and influencers, including individual pieces of work in which it played a substantive role. Disclosures can be in writing or verbal, depending on the circumstances. Employees should also disclose the use of generative AI tools as part of the drafting or creation process.
- Promote diversity and inclusion by being aware of biases incorporated in AI-generated prompts and output, using diverse perspectives within the agency to review prompts and content created by generative AI tools to ensure no bias is overlooked or shared externally.
- Avoid relying on generative AI tools to translate or create documents into other languages, and do not use generative AI as a replacement for diverse experiences, insights, or engagement. Finally, do not use generative AI tools to create imagery, likenesses, or avatars that create discriminatory content.
