One of the worst things about AI is how it validates my intelligence in plausible ways. I’ll reject an idea it comes up with and offer one of my own, and it tells me “your idea is much better, here’s why,” with specific details about my brilliance. Who wouldn’t want to believe it? The fact that the compliment isn’t coming from a being with any experience in the real world, and is actually from a computer that’s tuned toward sycophancy can easily slip my mind.

I’m also guilty of the opposite way of thinking. After I posted my AI disclosure, I reviewed my work and realized that there’s actually quite a bit that I’ve written entirely by hand, with just a quick check for typos, flow, and grammar. My disclosure might make it seem as though everything I do here is processed by AI, but that’s not true. So why do I feel defensive? Why am I uneasy about this stuff?

I think part of it is because I’m worried that confirmation bias is clouding my judgment. Lots of people I admire and respect believe that any AI use is wrong, and lots of the pro-AI voices I hear are coming from people that I don’t like all that much. I look around the virtual room and ask myself, “are these grifters and fanatics my people now?” I desperately hope not. But there may be something important I’m missing, and the more I use AI, the more concerns I have about how things will turn out.

I’m also the type of person who can hear a persuasive argument and agree in the moment, but after I take some time to process what I’ve heard, I often change my mind. If I’m being kind, I’ll say it’s because I like to keep myself open to ideas, and going in trying to poke holes prevents me from fully absorbing what’s being presented. If I’m being real, I might admit that I can be a bit gullible, at least at first.

Another part of it is the extreme level of hatred and vitriol directed against people who use AI. I recently read Ed Zitron’s enormous AI Hater’s Manifesto, and he brings up so many good points about the industry and technology that I wish more people could engage with his ideas. It’s filled with so much profanity and hatred, however, that it’s virtually unusable as a resource. Outrage sells, it seems, but who does it help? It’s one thing to rail against corporations and billionaires, but what about ordinary people who might not know better? If I’m the owner of a local restaurant and use AI slop to advertise my food, could it be because image generation tools are being shoved in my face everywhere and I look at my competition and feel as though stock photos are so 2022? Could I actually be a decent, hard working person with a million things to do who just didn’t think it through?

Thankfully, there are people who seem to share my thinking about AI as normal technology. Not a superintelligence, not something that will engulf the planet and extinguish life as we know it. A tool, with us humans deciding how it will be used. There are great things that are being done with it, right now, as well as all of the horrible crap that gets all of the attention. Its ubiquity is not inevitable. We’re not too late to make sure it works for us instead of against us.

I truly enjoy using AI, even as I worry about developing dependence and weakening my ability to think for myself. I have done amazing things with it I would never have attempted before, and I try to be very deliberate about how I use it. Is my way the right way? For me, for now, yes. I’m sure it’s going to evolve over time, but here’s where my personal AI policy stands at the moment:

Tom’s Personal AI Policy

  • Monitor usage. Limit the duration and days of use. Take time off, and don’t use it for boredom or companionship.
  • Be mindful. Is using AI the best choice? How significant will its impact be? Will it harm your ability to think and act for yourself?
  • Be skeptical. Don’t automatically trust outputs, and verify any claims before using them.
  • Be accountable. Own what you do with AI, and accept responsibility for your use of it.
  • Make sure it’s worth it. The resources expended to use these tools need to result in something significant enough to justify their expenditure.
  • Push yourself. Beyond simple efficiencies, use AI to enhance learning, extend thinking, and expand boundaries.
  • Share. Take what you learn and share it to help others navigate this new world for themselves.

This helps me feel as though I’m using these tools with thought, respect, and care. I’m still feeling defensive about it, though, and I’m still not quite sure why. Perhaps that’s a feature, not a bug. I wonder if everyone should consider sitting down a moment to create their own AI policy. What would yours look like?