Death by AI
One man's struggle with his mortality.
I found out about my death the way everybody finds out everything: from Google.
What happened was, I Googled my name ("Dave Barry") and what popped up was something called “Google AI Overview.” This is a summary of the search results created by Artificial Intelligence, the revolutionary world-changing computer tool that has made it possible for college students to cheat more efficiently than ever before.
The Google AI Overview for my name consisted of some basic biographical information about me, along with a list of questions under the heading "People also ask." This was the first question:
As you might imagine I was curious to find out what happened to me, so I clicked on that question. This was the answer:
Give Google AI credit for what it got right: That is, in fact, a picture of me, and I did, in fact, win a Pulitzer Prize (trust me, I'm just as shocked as you are). But to the best of my knowledge, I did not pass away last November 20. That is not just my opinion. In recent months I have been examined by two different licensed physicians, and if I had been dead, I'm pretty sure at least one of them would have mentioned it. ("Dave, your pulse is zero, and your blood pressure is zero over zero. I'm going to try your other arm, but frankly at this point I'm concerned.")
Fortunately, Google AI Overview has a mechanism whereby if you have an issue with something, you can submit feedback. So I submitted this:
I'll be honest: When I submitted this feedback, I did not expect much in the way of a response. Google is a huge company with a vast network of computers processing more than a billion trilobites of data every second, which is all the more impressive when we consider the fact that a trilobite is a marine creature that became extinct 250 million years ago, which I think we can all agree is a good thing because it looked like a giant cockroach with way too many legs:
So I didn't expect that Google, what with all it has on its plate, would have time to respond to my issue about not being dead. But I was wrong. Shortly after I submitted my feedback, the "People also ask" part of my AI Overview was changed. That was the good news. The bad news was, it was changed to this:
Yes. Google AI Overview kept the only statement I objected to — specifically, that I was dead — and removed the accurate statements. It apparently replaced these with information about a completely different Dave Barry, a political activist from Dorchester who died in 2016.
For the record, it sounds as though that Dave Barry was a much better person than I am. He fought for what he believed was right, whereas the only principle I ever have really stood for, over the course of my journalism career, is that Americans should not be required to use low-flow toilets. I'm confident that the late Dave Barry from Dorchester would not have wanted to be mistaken for me.
So I started to write another feedback message to Google AI Overview. While I was doing this, a little box popped up on my screen asking me if I wanted to chat about my issue. So I did that. Here's a transcript of the chat (I am not making this up):
GOOGLE AI: What can I help with? A brief explanation is all I need.
ME: I'm Dave Barry. The Google AI Overview for me says that I'm dead, and that when I was alive I was a political activist in Dorchester. Neither of these things is true. I am not dead, and I am not a political activist in Dorchester.
GOOGLE AI: Sorry, I didn’t understand your question. Can you try rephrasing it? Short phrases work best.
ME: I'm Dave Barry. I am alive. Google AI says I am dead. I am not dead. Google AI also says I was a political activist in Dorchester. I was not. I still am not.
GOOGLE AI: Apologies, I’m still having trouble understanding you. Let’s give it one more try. Please describe your issue in the text box below.
ME: My issue is that Google AI says I am dead, but I am not.
Google AI responded to that with this:
That's when I gave up on the chat. It was like trying to communicate with a toaster. Actually, it was worse than that. I bet if I stood in front of a toaster and shouted "I'M NOT DEAD" at it enough times, eventually it would respond "Oh! I get it now! You're not dead!" Whereas Google AI seemed incapable of grasping this point.
Nevertheless I gave it one last shot. I submitted the following updated feedback to Google AI Overview:
At that point I didn’t have much hope. But as the old saying goes, "Just when you least expect it, the unexpected happens." And sure enough, the next day the Google AI Overview about me had been changed again:
For the record, the second and third sentences are inaccurate: I've had two books1 published since Lessons from Lucy (which was published in 2018, not 2017) and I haven't written a regular column for the Miami Herald in 20 years, and when I did it was weekly, not monthly. But I didn’t want to get nitpicky just because the answer was still mostly wrong. The important thing was, Google AI had decided that I was no longer dead, and for that I was grateful.
I felt pretty good about this for nearly two days. Then yesterday morning I checked my Google AI overview again to see if anything new had happened to me, and saw this:
So after my miraculous recovery, I was dead again. But guess what? This morning, as I was getting ready to post this essay on Substack, I checked my Google AI Overview one last time, and:
“There seems to be some confusion” states Google AI. Yes! There does! And the source of this confusion appears to be… Google AI!
Anyway, the good news is that, as of right now, according to my Overview, I am once again alive. But I’m not making any long-term plans. I’m not even making lunch plans.
What can we learn from my experience? We can learn that although Artificial Intelligence is an awesomely powerful tool that according to experts is going to completely transform human existence, it is not very bright. So for now we probably should use it only for tasks where facts are not important, such as writing letters of recommendation and formulating government policy. We should NOT let AI handle jobs requiring a high degree of accuracy, such as airplane navigation. ("Ladies and gentlemen, we're beginning our descent into Honolulu, and we should soon... No, wait, sorry, this appears to be Paris.")
At least that's what I think. Now it's time to find out what you always-helpful paying subscribers think. Today we have two polls:
Swamp Story (2023) and Class Clown (2025). Both of these books are for sale in exchange for money.
“I’m sorry, Dave, I’m afraid I can’t do that.”
A writer friend of mine said you need to think of AI as Artificial Intern and check everything it says. It, like me (a writer of fiction), makes stuff up. Now I have to go check to see if I'm alive.