One Big AI Idea: Rewrite for tone

Tone is such an underrated part of writing. It’s what makes the difference between a piece that’s fit for the boardroom, for employees of diverse reading ability or for a jokey TikTok.

In many cases, the content of our messages doesn’t change much. We still need to convey the same information. But the way the words are put together can drastically change how they’re received.

This is an application where AI can shine. It often performs best when it’s working from something you’ve already done and merely rearranging rather than having to create something from whole cloth. Rather than having to rewrite over and over again, write once and let AI tweak the tone. Of course, you’ll have to come back  for a thorough edit, but it can greatly reduce your tinkering time.

Here’s an example. We gave ChatGPT a paragraph from a filing from Costco’s board rejecting a proposal about DEI initiatives. Here’s the original copy:

Consistent with our commitment to “obey the law,” we regularly evaluate our practices concerning compliance with law, including evolving Supreme Court decisions. We believe that our diversity, equity and inclusion efforts are legally appropriate, and nothing in the proposal demonstrates otherwise. As part of our obeying the law, all decisions regarding recruiting, hiring, promotion, assignment, training, termination, and other terms and conditions of employment will be made without unlawful discrimination on the basis of race, color, national origin, ancestry, sex, sexual orientation, gender identity or expression, religion, age, pregnancy, disability, work-related injury, covered military or veteran status, political ideology or expression, genetic information, marital status, or other protected status.

Fairly dense stuff. Let’s say you want to explain this in plainer, more casual language for front-line employees. Here’s what ChatGPT gave us.

We want to let you know about something that’s important to us: treating everyone fairly and following the rules. We regularly check how we do things to make sure we’re sticking to the law, even when rules or decisions change. We believe our diversity, equity, and inclusion efforts are on the right track, and there’s nothing that shows otherwise.

This means that whenever we make decisions about hiring, promotions, training, or anything else related to your job, we’re committed to being fair. We don’t treat anyone differently because of their race, where they’re from, gender, religion, age, disability, or any other protected characteristic. Everyone here deserves to feel respected and valued, and we’re making sure that happens.

Or let’s say you want to approach this in a more playful way, perhaps a short-form video hitting back against the criticism you’ve been getting. We asked ChatGPT to make it humorous.

We’re all about following the law—no shortcuts, no bending the rules, no sneaky stuff. We regularly check our practices to make sure we’re in sync with the latest Supreme Court drama. We’re confident our diversity, equity, and inclusion efforts are on the up and up, and nothing in the proposal has raised any red flags.

When it comes to hiring, promotions, training, and all that work-related jazz, we don’t play favorites. No matter your race, color, where you’re from, who you love, how you identify, how old you are, or whether you’ve got a pet iguana named Steve, we’ve got you covered. We’re an equal-opportunity operation, and we wouldn’t have it any other way!

Is it hilarious? Well, comedy is subjective. One last example. Perhaps it’s Take Your Child to Work Day and you need to explain this to a group of 5-year-olds. Here’s how ChatGPT broke it down:

We always follow the rules because that’s the right thing to do. We check what we do to make sure we’re being fair and following the law. We believe it’s important to treat everyone equally and include everyone, and we think we’re doing it the right way.

When we make decisions about things like hiring people, giving promotions, or teaching new skills, we promise to treat everyone fairly. We won’t treat anyone differently because of things like their skin color, where they’re from, if they’re a boy or girl, what they believe in, how old they are, if they have a disability, or anything else that’s protected by the rules. We want to make sure everyone feels safe and respected.

One piece of content, consistent information, rewritten for three audiences in as many minutes.

How have you used ChatGPT to alter tone?

For more AI insights, joins us for the first AI Horizons Conference in Miami, Feb. 24-26. 

Allison Carter is editorial director of PR Daily and Ragan.com. Follow her on LinkedIn.