The Communications Your Firm Should Never Let AI Handle (And the Ones It Should)
Would you let just anyone on your team respond to an angry client? Of course not. Dealing with an angry client needs an experienced member of your team, someone with the authority to respond appropriately and do what’s necessary to resolve the issue.
Then why would you let AI respond to an angry client?
What AI Should Never Touch
Crisis communications. Full stop.
When a client is dealing with a problem, when stakeholders are upset, when something has gone sideways: AI cannot handle it.
Why?
Because AI doesn’t understand context. It can’t read between the lines. It can’t sense that someone is hurt or anxious or one bad response away from walking out of the door.
Human judgment is non-negotiable here.
Same goes for client complaints.
If a client says “I’m not happy with this,” your response better come from a real person who understands what’s actually happening, not AI generating something that sounds polite but misses the point entirely.
Media interviews and thought leadership are another hard no.
AI can help you research. It can organize talking points. But the insights that make you quotable? Those have to come from your knowledge, your experience.
Otherwise, you’re just another voice saying what everyone else is saying.
Where AI Actually Helps
Now don’t get me wrong, there are plenty of places AI does solid work.
Media monitoring? Let AI track coverage and flag what matters. Then a human reviews and decides what to do about it.
Routine status updates? Fine. AI can compile information into a first draft. You review and refine.
Meeting recaps, event summaries, research compilation; all fair game for AI assistance.
The key word: assistance.
AI drafts. Humans review, edit, and own the final product.
The Test You Should Use
Before you hand something to AI, ask three questions:
- Does this require understanding and context AI doesn’t have?
- Could getting this wrong damage a relationship or your reputation?
- Does this need professional judgment or ethical consideration?
If you answer yes to any of those: AI stays in the background.
Building Team Guidelines
Don’t just tell your team “use good judgment.”
Give them specific examples.
“You can use AI to research industry trends for a client presentation – then verify and analyze the findings yourself.”
“You cannot use AI to respond to a client concern without reviewing and editing the response personally.”
Make it concrete. Make it easy to follow.
Why This Matters
Professional services run on trust.
And trust gets built in moments – especially the hard moments.
When a client is frustrated and you respond with something that feels automated or generic? You just told them they’re not worth your personal attention.
When you show up in those moments as an actual human who understands their situation? That’s when relationships get stronger.
The Bottom Line
AI can make your firm more efficient, but it cannot make your firm more trustworthy. And in professional services, trust is the product.
So use AI for the work that doesn’t require relationship judgment and keep humans in charge of everything that does. Because your clients aren’t paying for perfectly formatted documents – they’re paying for someone who understands their situation and gives them advice they can trust.
AI can’t do that.
You can.
Need help developing your firm’s AI strategy?
HMA Public Relations works with professional services firms to create practical AI policies, train teams on smart AI use, and integrate technology without losing the human judgment that makes your firm valuable. We’ve been navigating communications technology since fax machines were the rage and phones were actually used to make calls; we can help you get this right.
Contact me at afink@hmapr.com, 602-957-8881 or visit hmapr.com.
