I want to be honest with you about something most candidates, and politicians, avoid talking about directly. I do use AI. I also have real concerns about it. Both of those things are true, and I don’t think people are well served by pretending otherwise.
Running a campaign, a small business, or a political office on today’s timelines means doing the work of what used to take three or four people. AI helps me draft faster, research faster, and respond faster. It also helps me to see the things that I might not have naturally… Though, as I write this, I think about the stories of AI psychosis. Clearly, there are times that people spiral in an AI echo-chamber rather than having the same experience I do.
I guess the bottom line is that I’m not going to pretend I’ve opted out of a tool that nearly everyone competing for attention and resources is already using. Refusing to use it wouldn’t be principle. It would just be losing.
Using a tool doesn’t mean I’m at peace with what it’s doing to the world.
Environmentally, the data centers powering these systems use water and electricity at a scale people never see reflected on their own bill, often in communities that had no say in the trade-off. Efficiency for me can mean real cost somewhere else, carried by someone else. I don’t think that cost is being tracked honestly, and I’m not going to pretend I’ve settled this just because I’ve found the tool useful.
Societally, I’m worried about something harder to see: the disappearance of entry points & the ‘cognitive offload’. The junior researcher job, the first-draft writer, the entry-level analyst. These roles were never just cost centers. They were how people became senior researchers, editors, and experts. When a company replaces one of these roles with a tool that produces something passable, it hasn’t just automated the job. It has removed the training ground for the next generation of people who would have grown into it. A passable memo is not a lawyer. A passable draft is not a writer. What’s being substituted here is often a profession, not a task, and that distinction matters more than most people are willing to admit.
The ‘cognitive offload’ is more related to “ChatGPT told me…” “Claude said…” statements we hear (or will hear) from friends relating to the simplest of things.
Our politics wants this to be binary. Either you’re standing in the way of progress, or you think AI is the best thing since sliced bread. I think that the binary is part of the problem, because it lets everyone avoid the harder governing questions:
I don’t have these answers worked out yet. What I can commit to is not pretending the tension doesn’t exist, and using my own use of these tools as a reason to wave away their costs.
