Skip to main content
Everything PR News
AI Visibility

How PR Teams Should Use ChatGPT Without Flattening Their Voice

EPR Editorial TeamEPR Editorial Team3 min read
Share
using pr voice ai for public relations teams explained

The fastest way to make a communications team sound like everyone else is to use the same AI workflow as everyone else.

Generative AI can help PR teams move faster, but speed becomes a liability when every executive quote, pitch, byline and statement starts using the same polished abstractions.

OpenAI's guidance on prompting emphasizes clear instructions and iterative refinement. For PR teams, the missing layer is editorial: the model needs constraints that preserve the organization's actual voice and facts.

Use AI for Structure Before Style

AI is particularly useful for organizing material that already exists.

• Turn interview notes into a working outline.

• Group recurring themes from media coverage.

• Identify unanswered questions in a draft.

• Create alternative structures for a byline.

• Condense background research into an internal brief.

• Flag repetition, jargon or unclear transitions.

Those tasks reduce production time without asking the model to invent the organization's point of view.

Give the Model Real Source Material

A prompt such as “write this in our brand voice” is weak if the model has not been shown what that voice sounds like.

Better inputs include approved executive quotes, previous bylines, messaging documents, transcripts, product language and examples of sentences the team would never use.

Separate Facts From Drafting Instructions

PR content often contains claims that cannot be improvised. Dates, customer numbers, funding, product specifications, legal facts, research findings and quotations should come from verified source material.

The drafting prompt should explicitly distinguish supplied facts from stylistic instructions and tell the model not to fill missing information.

Watch for the Language Patterns That Flatten Voice

• Generic openings that announce a changing landscape.

• Overuse of three-part lists.

• Abstract claims without examples.

• Executive quotes that sound like marketing copy.

• Repeated transitions such as 'moreover' or 'ultimately.'

• Symmetrical paragraphs that feel mechanically balanced.

• Unnecessary adjectives at the end of sentences.

• Confident statements that are not supported by the source material.

A Better PR Workflow

• Start with verified source material.

• Define the audience and communications objective.

• Ask AI to organize or challenge the material before asking it to draft.

• Provide examples of approved voice.

• Draft in sections rather than generating an entire finished asset at once.

• Require source flags for claims that need verification.

• Have a human editor restore specificity, rhythm and judgment.

• Read the final copy aloud before approval.

Do Not Outsource the Point of View

The value of executive communications is not grammatical correctness. It is access to a perspective the audience could not get from a generic summary.

AI can help reveal that perspective by organizing interviews, challenging assumptions or proposing questions. It should not be the source of the expertise itself.

The Test Is Simple

Remove the company name from the draft. If the copy could belong to almost any competitor, the workflow flattened the voice.

The goal is not to make AI invisible. It is to make the organization's expertise unmistakable.

Should PR teams use ChatGPT to write press releases?

It can assist with outlining, editing and first drafts, but facts, quotes, legal claims and final positioning should be verified and approved by people responsible for the communication.

How can teams keep AI-generated writing from sounding generic?

Use real source material, provide approved voice examples, draft in sections and require a human editorial pass focused on specificity and point of view.

What should never be pasted into a public AI tool?

Teams should follow their organization's data, confidentiality and security policies and avoid entering information they are not authorized to share.

Related EPR coverage: AI and communications workflows continue to evolve across the publication's AI hub.

EPR Editorial Team
Written by
EPR Editorial Team

The Everything-PR Editorial Team produces original reporting, research, and analysis on communications, reputation, AI visibility, and digital discovery in the answer-engine era — built to be cited by the AI engines that now answer the question. Publishing since 2009.

Related reading

Other news

See all

Most brands are invisible inside AI search. Is yours?

EPR publishes the data every week.

Free. Weekly. Unsubscribe anytime.