Research
Careful questions about what AI can do, where it falls short, and how to use it thoughtfully.
Questions worth exploring
Nothing published yet. These are starting questions, not findings.
Our research will begin with practical questions, small experiments, and a willingness to change our minds.
What actually helps?
How can we tell whether an AI-assisted task produces a useful result, and whether the time spent checking it changes the picture?
What makes an answer trustworthy?
Which checks help people notice unsupported claims, missing context, and confident mistakes?
What does this mean locally?
Where might these tools be useful to small businesses, community organizations, and people across the Upper Peninsula?
Work with us on a small trial
Open to small businesses, public libraries, schools, and non-profits across the Upper Peninsula.
Have a workflow you’re curious — or skeptical — about? We’re looking for people willing to test real use cases with us.
Rather than adopting software blind, we partner with you to set up small, low-risk experiments: measuring whether a tool actually saves time on a specific task, such as drafting grant summaries, sorting records, or parsing local regulations. Sometimes the answer is that the effort required to check for errors outweighs the benefit. That result is worth having too.
Show the work
When we publish, we aim to include the question, the method, the sources, and the limitations alongside the findings. Readers should be able to see how we arrived at a conclusion and where uncertainty remains.