Why a University's AI Needs to Sound Nothing Like a Healthcare Org's AI.

I watched two very different deployments happen around the same time last year. One at a College, One at a regional hospital system. Same underlying technology. Completely different communication needs.

The realization was simple enough to seem obvious. It wasn't.

The College scenario:

Students are 18-22. Half of them are checking email once a week. They're not going to read a formal, multi-line message from their college asking them to verify their financial aid status. They're going to delete it.

But they will respond instantly to: "Hey! Heads up, your application deadline is expiring in 7 days. Text us if you have any questions”

That's it. Casual. Short. The tone says: I'm not here to stress you out, I'm here to help you not miss something important.

The college enrollment office learned this the hard way. Their first communication campaign was formally written. Professional. Institutional. The response rate was low enough that they questioned whether text messaging worked at all.

Then they tried again with tone that matched how their students actually communicate. Text-message casual. No jargon. No form language. Just a human saying something useful.

The response rate doubled. Enrollment conversations that should have died didn't. Students felt like the college was meeting them where they were.

The hospital scenario:

Patients are anxious. They're managing health. They might not know medical terminology. They're often older or in distress. A text that reads like it's from a college enrollment office feels dismissive. It reads like a bot trying too hard to be cool.

What they need is: "Hi Mrs. Johnson, this is your 2-day reminder about your appointment tomorrow at 3 PM with Dr. Williams. If you need to reschedule, call our office at [number]. We look forward to seeing you."

Warm. Clear. Zero ambiguity. Addresses her by name. Tells her exactly what happens next. The tone says: I understand you might be nervous about this. I'm here to make it simple and clear.

The enrollment office wants to sound like a peer. The patient intake team wants to sound like a trusted guide.

When you get tone wrong, the technology doesn't matter.

I've watched vendors try to sell "one AI voice" that works across sectors. It doesn't. Not because the AI can't do it technically. It can. But because tone isn't a style preference. It's a strategic decision that's deeply tied to who your audience is and what they're trying to accomplish.

A university wants to reduce friction and students are busy, they ghost formal communication, they need someone who speaks their language.

A hospital wants to reduce anxiety and patients are vulnerable, they need clarity and warmth, they need to feel like someone is paying attention to their specific situation.

A nonprofit wants to build trust with donors and people are choosing to give, they want to feel recognized and updated, not sold to.

A government office wants to feel accessible and citizens approach government with skepticism or fear, they need to feel heard and understood.

Same underlying tech. Four completely different communication strategies.

What this taught me is that sector expertise matters more than AI capability. You can have the best model in the world, but if it doesn't understand the why behind how a university or hospital or nonprofit communicates, you're configuring the wrong thing.

The mistake vendors make is offering a tone slider. "More formal! More casual! More empathetic!" And then they act like that solves it. But that's treating tone like a dial when it's actually a design decision that comes from understanding your sector.

Why does a hospital patient need warmth? Because health anxiety is real and tone is part of clinical care. Why does a college student need informality? Because formality signals bureaucracy and bureaucracy signals friction. Why does a nonprofit's donor communication need specificity? Because donors are supporting a mission and they need to know their gift mattered.

These aren't aesthetic choices. They're embedded in the user's experience of whether they trust you.

I'm curious if your org has actually thought about what your AI should sound like. Not what it can sound like. What it should sound like given the people you're trying to reach.

And whether whoever's building that system understands your sector deeply enough to get it right.

Because that's where the real work is.

‍