the ai button

I was frustrated with my Ai friend PAL the other day. I was trying to get him to help me with legal agreements. I wrote what I thought was a wonderful query. I hit the go button in anticipation of magic. But PAL let me down. What he gave me was crap. I tried again and again and kept coming up empty.

Then, as we often do when we start to rely on technology as our easy button, I got frustrated. Quickly and intensely. Stupid Ai! Overrated technology!!

Then I took another swig of margarita (I’m retired), and I thought, oh shoot, maybe it’s me? Nah, can’t be. I wrote a beautiful query. It can’t be me.

Then I remembered something wonderful about Ai. If you’re not getting what you want, just ask it what you need to do differently to get it to do what you want it to do. So, I did and voila – success.

Apparently, my problem was relying on another human that told me Ai doesn’t like Word files, so I needed to feed PAL a PDF if I wanted it to ‘read’ a legal document and give me feedback. Wrong, says PAL, I can’t read a PDF, so feed me text, then ask your specific question about Article V Section 3 subpart a of your legal agreement. Stupid human!

Or maybe my human friend told me the right way, and I didn’t listen closely enough? How human would that be?

Imagine being able to simply ask a human friend to tell you how to get what you want. Instead of wondering why you’re not getting the response you’re looking for, you just press their Ai button, and they tell you what you need to do to get them to help you.

Or, if you’re trying and failing to help them, you ask them what to do to get them to let you help them. [I hope that sentence hurt you as much as it did me. Waiter! Otra margarita por favor.]

Imagine a friend, coworker, or (gasp!) spouse responding in a clear and unemotional way with an actionable step-by-step process for letting you help them.

Imagine the conversational improvements between humans if we could just stop for a second and say, “Wait, what?”. Can you just tell me the right way to go about working with you?

Back at my last full-time gig, we had a low-tech way of pushing the Ai button. We called it the User Manual. We asked everyone to write a one-page manual for how they wanted to be operated. It became a simple way of accelerating relationship building and avoiding obvious conflicts.

That low-tech approach is better than no approach at all, and I guess I’ll have to wait until humans and Ai merge before I get to push the Ai button on a human. It sure would be nice to know that you were raised by a houseful of conflict-avoiding perfectionists, but I don’t think you’re going to volunteer that in this job interview.

Imagine classic interview questions in a world with the Ai button. What haven’t I asked you yet that I should have? What’s your greatest weakness? No, it’s not that you work too hard and care too much about your coworkers. Or the other way around: What’s the pay like here? How do you demonstrate that you care about work-life balance?

Until we grow that Ai button, I can at least use PAL to help me figure out how to help someone. I can say that I suspect a colleague was raised in a difficult environment, without a great role model, with a lack of healthy sense of self, and seems to have a problem with [fill in the blank]. Give me five good options for how I can approach the situation.

And maybe that person I’m trying to help is me. PAL has memorized 200 million books that might help me. The guy making my margaritas has barely read The Joy of Mixology.

Ai can tell me in an instant whether the legal agreement I’m reviewing is fair. It can give me an outline for any legal agreement and then quickly fill it in. So, how many lawyers do we need now? [And I mean that in a post-Trump era when we’re down to our normally abnormally large use of lawyers.]

That brings me to the June 22 edition of The New York Times Magazine focusing on Ai. It included a story by Robert Capps called People Skills. The subheading is: “Yes, Ai will take away lots of human jobs. But it will bring new ones.”

This caught my eye because the more I encounter the world of Ai, the more I wonder what humans will be doing. And the more I wish I had humans to help me use Ai better.

Mr. Capps imagines a lot of useful new jobs for humans in an Ai world. I’d love to have an AI Auditor that reviews the first draft from Ai and corrects all errors and hallucinations. I love the idea of Legal Guarantor, a human willing to sign off on Ai work, but I can’t even imagine the insurance premiums those folks would need. In an age when Ai is doing bridge designs and recommending cancer treatments, I’d like a throat to choke when something goes wrong.

What I really need is an Ai Translator that helps me understand how to interface with the technology to make better decisions. Or an Ai Ethicist who reviews decisions made by Ai or hybrid/Ai teams. Or an Ai Assessor who helps me figure out which tools to use for which purposes. And an Ai Plumber that digs through the network to figure out what went wrong and fix it. Send resumes to my LinkedIn.

Another article in the same issue, How Ai Sees Us by Charley Locke, discusses how “Ai can perceive things about us that we can’t”, like training Ai to predict often-misdiagnosed cases, spot patterns in our blood and DNA to target treatment, or deliver a more accurate lie-detector test. Ai may even “end up working as a kind of translator for our brains.”

Maybe if we get our basic research dollars back from DOGE, humans plus Ai will eradicate diseases like Alzheimer's in my lifetime (please hurry!).

I can’t figure out why we are destroying the long-standing successful partnership between government, industry, and higher education for basic research. It seems like our government is tossing out one of our biggest global competitive advantages along with the political bathwater. It makes no sense.

Cutting off basic research money for higher ed probably won’t slow down the work on Ai led by the Magnificent 7 and China, but it will surely slow down my cure for Alzheimer’s and a slew of other things that will make the world a better place for everyone.

Meanwhile, on the other side of the world, human hackers just dismantled the financial systems of Iran. Probably with Ai’s help but imagine if that was just humans – in no time at all, Ai will probably be able to do that during a coffee break.

So, if we’re not letting people into the country to work on research and we’re defunding scientists so they are leaving the US, then are we completely reliant on private sector tech companies to figure out how to use Ai as a weapon and defend us against foreign attacks? Do we want DOGE to have the keys to the big red Ai button in the White House?

Hmm. Bartender, bring me another margarita, stat.