In most sectors, AI is already part of the toolkit. It’s helping teams cut admin, spot trends, and connect systems that never used to talk to each other.

In homelessness services, we’ve barely begun.

Even with all the pressure of rising need, shrinking budgets and overworked teams, we’re still doing so much manually. Paper notes. Spreadsheets. Phone calls just to check if a service is still running.

So, what’s stopping us?

This is people-first work. That doesn’t change.

Trust, empathy, and human connection are at the heart of everything this sector does. You can’t automate that.

There’s a real concern that bringing in tech could dilute those relationships, or worse, replace them. And after years of digital tools that looked promising but didn’t stick, people are right to be wary.

But AI doesn’t have to get in the way. Done right, it can help. It can take pressure off, free up time, and give people more space to focus on the things that only humans can and should do.

Too many tools have promised a lot and delivered little.

Frontline teams don’t have time to experiment. If something takes too long to learn or doesn’t fit how people already work, it won’t last.

In this sector, we’ve seen that happen too often. So, caution makes sense.

Trust takes time. A good tool needs to prove itself. It should be useful, reliable, and make life easier, not harder.

Ethics and privacy come first.

When you’re supporting people who are already at risk, protecting their information isn’t just about compliance. It’s about care.

We need to be clear about what’s collected, why it matters, and who it’s for. There’s no room for grey areas. People need to know the system is working for them, not watching them.

At Street Support Network, we’re focused on practical uses of AI that reduce admin, not replace human connection. We’re exploring tools that help with strategic planning, marketing, fundraising, and day-to-day management—spaces where automation can free up capacity, not erode relationships.

Having led homelessness services myself, I understand the pressure teams are under. That’s why I’ve been learning to use tools like IBM watsonx, to build systems that take some of the pressure off and make our day-to-day work a little easier.

This isn’t about following trends. It’s about staying open.

If AI can ease pressure, save time, and help services connect better, then the question shouldn’t be if we use it. It should be how, and on whose terms.

Over the next few months, we’ll share more of what we’re learning. The ideas we’re testing. The honest bits about what’s working and what’s not. We’re also planning a network event to dig deeper into what this could look like across the sector.