Tri-State has a way of getting your brain going. Same thing this year.
AI came up a lot. It felt thoughtful. Pretty grounded. Still early for some of the real-world use.
“Now What” - Nelson Strickland (2026)
One session I went to was about using Claude as an AI agent.
Good pace. Real examples. Not hype.
It also brought up a question I’m still sitting with.
Most camps aren’t running high-end systems.
So if we’re going to use this stuff day to day, we’ve got to think about performance, access, and long-term usability too.
<aside> 💡
Core idea
Check with your organization before using AI with any sensitive information.
</aside>
It’s not flashy. But it’s the right place to start.
This thought comes from the speaker in the Claude AI agent session. His strongest stance in my opinion, and one that was received as maybe not well, but definitely expected.
Before you test tools, make sure your team is on the same page about what’s okay to share, store, or run through AI.
People are looking for answers. And we, the human collective, simply do not have them…Sway 🎹
<aside> 💡
Core idea
AI should support your instincts, not replace them.
</aside>
If you’ve been in this work a while, you know the feeling.
You can tell when a staff member is going to struggle.
You can tell when a program needs to shift.
You can tell when the day is about to go sideways.
“We don’t question cars, spell check, or Google. But when AI saves us time? Suddenly it feels like we’ve skipped a step. Maybe the problem isn’t the tool. Maybe it’s the belief that “real work” has to be hard.” -Melissa Lloyd
AI doesn’t replace that.
It helps you sort your thoughts.
It helps you pressure-test what you already see.
It helps you move faster once you already know.
