When the AI Prompt Goes Wrong

AI can write faster than most humans—but not always better. In communications, where tone and nuance matter as much as grammar, a single off-prompt can turn a heartfelt message into something hollow or oddly robotic. “When the prompt goes wrong,” your message might technically say the right thing, but emotionally miss the mark.

Still, when used well, AI can be an incredible partner for communicators. It can help us brainstorm new angles, find the perfect tone for different audiences, and even stress-test how a message might land across teams. The key is learning how to prompt like a communicator, not a coder.

Here are four practical ways to get AI prompts right:

1. Write like you brief a teammate.

Instead of vague prompts (“Write an internal email about a new initiative”), give context: who it’s for, what they care about, and how you want them to feel. AI mirrors your clarity.

2. Feed it your brand or culture voice.

Share a few sample messages that sound “right.” Ask AI to analyze their tone before it starts writing. That calibration step ensures output that feels human, not canned.

3. Ask for options, not perfection.

Use AI as your brainstorming partner, not your ghostwriter. Ask for three tone variations (“professional, conversational, heartfelt”) and mix what works. The best version is often somewhere in between.

4. Test for emotional accuracy.

Before sending, ask: How might this message make someone feel? Then prompt AI to rewrite for empathy or motivation. It’s a powerful gut-check for tone.

AI won’t replace communicators—it will amplify them. The difference between awkward and authentic often comes down to the prompt. When it goes wrong, the fix isn’t more polish—it’s more humanity.

To help you think about where AI can make the biggest difference in your daily work, here’s a quick guide to how it can elevate different types of communications: