The PR Council Updates Ethical Guidelines on Generative AI
The PR Council has updated its "Guidelines on Generative AI Tools," offering a more comprehensive framework for the ethical use of generative AI in public relations, an overview of the current state of AI, common use cases across agencies, and guidance on disclosure to clients, employee training, and more.
Since the initial publication of the guidelines last April, the landscape of generative AI in the communications industry has evolved significantly, with some firms moving from cautious experimentation to tactical implementation and even strategic adoption.
"While core ethics principles should not change, actionable guidelines need to evolve with the industry. One of the areas of rapid AI evolution has been developments with generative AI in enterprise software, such as Adobe's Firefly, and how agencies are using generative AI to create visual outputs. We convened a focus group with agency creative directors along with a cross-disciplinary team of experts in ethics, corporate reputation, digital, and DEI (diversity, equity, and inclusion) to update the guidance to deal with today's reality," said Mark McClennan, general manager of C+C and Ethical Voices podcast creator.
The new guidelines introduce a continuum that outlines the evolution of AI and its implications in serving clients, supporting teams, and advancing the public interest.
"We are hearing more optimism about how these [AI] tools, when paired with human ingenuity, can improve our work and eliminate certain mundane or time-consuming tasks," said Anne Green, CEO of G&S Business Communications, in a statement. "Augmenting human intelligence and ingenuity is the key. It is still difficult to foresee a time when these platforms and tools can produce unique and fully client-ready outputs totally free of issues with no human engagement or review."
In addition to the continuum, other updates to the PR Concil's AI guidelines include the following:
- Creatives embrace generative AI, especially enterprise tools such as Adobe';s Firefly and Getty Images' generative AI engine, for a range of use cases, but all agree none should be presented as final product to clients. Caution about copyright and trademark issues abound, and creatives agree that human genius is required to create unique, on-strategy outputs.
- Beyond the creative realm, AI capabilities are being rapidly added to many of the tools and platforms routinely leveraged by agency professionals. Keeping abreast of the new capabilities, challenges and biases of existing tool sets should be a top priority for all firms.
- More than ever before, agencies and clients need clearly articulated policies about the use of generative AI, including transparency requirements. This need extends to all vendors, including influencer or creator relationships.
- While the large language models are working to eliminate hallucinations (some experts predict this issue could be significantly improved or even eradicated within 18 months), proper sourcing is still necessary and remains a critical skill for all agency professionals. It is not enough to have a source; agencies need to check that source as well.
- The leading generative AI platforms report progress, but the PR Council continues to advise caution when inputting confidential client information, due to mistrust of promised security and confidentiality measures.
- The opportunity for bias is likely to always be an issue with the output of generative AI tools, so following a checklist to remove bias is important.
"As generative AI technology continues to advance at a rapid pace, it's crucial for firms to stay informed and implement policies accordingly," said Kim Sample, president of the PRC, in a statement. "Our updated guidelines are designed to empower PR professionals with the knowledge and tools they need to harness the potential of generative AI while being aware of the risks."