Ask anyone who works a difficult job what resource they wish they had. Chances are, they’ll say time.
My wife and I say this to each other constantly. How was your day? Ran out of time. Massive to-do list, never stopped moving, never stopped working, and still couldn’t get to everything. I’ve definitely felt that way about my work with the Foundation for Hao-Fountain Syndrome (usp7.org). I want to be in touch with more families. I want to get more people on Matrix, our patient registry platform, as part of our new initiative, Project Centauri. I want to do a million things more than I have time for.
So recently, I’ve been exploring what AI can do for us. And I’ve made some major headway.
The Chatbot That Didn’t Cost $30,000
For years, I’ve been asking developers to build me a chatbot. The ideal scenario: I draft a comprehensive document with answers to every question newly diagnosed families ask. Things like, what do I do first? Which doctors should I take my kid to? The chatbot would guide them through all of it without requiring a phone call or a chain of emails.
Every developer I talked to quoted $25,000 to $30,000. Then I presented the same question to an AI platform. About an hour later, I had a working chatbot. No developer needed.
It’s not quite ready for prime time yet, but early testing is promising. The chatbot is designed to shepherd newly diagnosed families through the process of joining our foundation, connecting with Dr. Christian Schaaf (the world’s expert on our gene and disease), and getting onto Matrix. And if things get complicated, it connects them to a human.
The Research Assistant I Never Had
I’ve always admired the behind-the-scenes assistant who reminds the boss about things falling through the cracks. So I told the AI about the various scientists working with us, our short-term goals, and what science we still need to nail down. Then I asked: who’s missing? Who should be on our team but isn’t?
The AI read through a pile of scientific papers and came back with names I’d never heard of. Scientists whose work would help us immensely. I’m no fool, so I didn’t start drafting emails right away. I asked to see the papers, read them myself, and found that the AI was right. These researchers would be an excellent fit. I’ve reached out to some of them and am waiting to hear back.
This function of AI as an assistant that can step back, look at things, and figure out what you’re missing is huge for a one-person operation like mine.
The Dashboard That Took Minutes, Not Weeks
Project Centauri has us tracking dozens of families at different stages: some on Matrix but without their genetic report uploaded, some with medical records but no genetic data, some not on Matrix at all. Tracking all of this manually would make your head spin.
So I had the AI build me a dashboard. I can see how many people are on Matrix, how many have completed each step, when they were invited, and what I need to do next. It’s a simple spreadsheet that would have taken me ages to build on my own. It took a few minutes.
These homemade tools, these methods of measuring our impact, of seeing what families are doing or not doing, it’s like having a bunch of new employees. Departments, really. The way we’ve always operated is that if we need a thing and it doesn’t exist, we have to make it ourselves. I’m glad that AI is making that possible without eating up entire days.