AI in PR: What we learned at UnSummit 2025

At PRSA Dallas’ UnSummit 2025, there was one panel that underscored what we’re seeing almost every day: AI in every industry, not just PR, is here to stay, but without policy, it’s a major liability.

The discussion revealed a gap that communicators can no longer afford to ignore. Many companies lack a clear, actionable AI policy, and even fewer make those policies public. That’s a missed opportunity. A public-facing policy builds trust, demonstrates accountability, and reduces the risk of an AI-related crisis.

We’ve already seen how quickly AI missteps can erode credibility. Some of the panelists’ examples included AI-generated campaign images criticized for lack of diversity, fake quotes slipping into published stories, or a summer reading list filled with books that don’t exist.

These aren’t just embarrassing mistakes, they’re operational failures. The instinct to “blame the AI” doesn’t hold up. Ultimately, humans are responsible for all outputs, no matter how they’re created.

Beyond reputational damage, there are deeper risks at play: data leakage, employees turning to unapproved shadow AI by using tools on their personal devices in the workplace, and vendors using AI without disclosure. Every one of these creates vulnerabilities that a strong policy can address.

The best AI policies don’t stifle innovation, they enable it responsibly. That means defining your organization’s red lines, making it clear what AI can and cannot do, from content generation to note-taking. It also means ensuring disclosure when AI is used, and equipping staff to spot bias, vet outputs, and use prompting techniques that minimize risk.

While this all seems scary in the moment, there are ways that AI can make our jobs more efficient when used responsibly, from ensuring your longform content is properly fed into the AI machines to get your message out, to supporting your day-to-day work to give you time to produce higher quality content. It’s time to think of AI as your assistant, rather than your eventual replacement.

As we reflect on UnSummit 2025, the lesson is clear: policy isn’t just paperwork, it’s protection. It’s how organizations turn AI from a potential crisis into a competitive advantage.