Every year over Thanksgiving weekend, I return to my childhood home, wear only my college sweatpants or skinny jeans from 2014, eat different variations of turkey for 3 days straight, and watch an egregious amount of TV with my family. And I always feel particularly grateful for the small joys that come with slowing life down a bit during this time.

This past Thanksgiving weekend, though, also marked 3 years since the release of ChatGPT, and I must say, I was feeling very thankful for how much easier my life has gotten with AI (ask me for all of my hacks and I will gladly share…).

If you know me and have spent any time with me in the last 3 years, you know I love to experiment with ChatGPT (I refer to it fondly as “Chat”). I’m of the opinion that we can’t fear the inevitable. AI is here and it’s not turning around, so I’d rather figure out for myself where it works and where it doesn’t.

So naturally, we were not going to have a Thanksgiving weekend without Chat coming up. But it came up in a way that surprised me.

When I come home and escape the constant motion of my life in NYC, I find that for once, I don’t have to feel bad about sitting on my couch binge-watching the newest TV shows because I’m doing it with my family. This time, we watched Pluribus, a bold, slightly unsettling series made by Breaking Bad creator Vince Gilligan.

Side note - if you’re reading this, you probably care about or are interested in AI, so you should really watch the show… I’ll do my best not to spoil any of the actual plot in the rest of this post.

So what is Pluribus about?

Honestly, for the first two episodes, I had no idea (and to be clear I still think this question is up for debate since new episodes are coming out weekly). I can tell that it’s one of those Severance-esque shows that kind of needs the entire season to bring it all together before the ‘aha’ moment. But after a few episodes, there were a few things that started to resonate with me in an eerie way…

TLDR: Pluribus follows Carol (Rhea Seehorn), one of only a few people who remain immune from a hive mind disease that has taken over humanity. She is a hater at heart, and her irritability stands out in stark contrast to the seemingly unbothered, happy nature of the hive mind also known as “them.” The show explores a series of brilliant but complex questions: if everyone knew everything about everyone, could we get over our differences and just… be happy? And if that is the case, and we are all one happy, shared mind (so no individualism), are we human at all?

Gilligan says that he started writing this show 5 years before ChatGPT even existed, but I can’t help but feel like Carol’s interactions with “them” are reflective of some of my interactions with Chat. There is a ton of discourse online about what Gilligan wanted this show to be about, but I’ll tell you what makes me watch from an AI commentary lens (and arguably, why you should too).

I think episode 4 sums it up pretty well. There is a scene in which Carol writes this list on her whiteboard:

What I Know About Them

1. Eager to please - would give me A-Bomb?!

2. Can’t kill, not a fly

3. Don’t play faves, like all jerks same

4. Trying to change me!

5. Weirdly honest?

Tell me this doesn’t sound ... familiar? We definitely have all complained about AI being a sycophant, generally a validator or cheerleader (even if you’re being an a**hole to it), and incapable of lying (yes, I know, it does get some facts wrong, but not maliciously).

Now, #4 is interesting in the context of AI. We could go down a rabbit hole and ask ourselves existential questions about influence and creativity like are my thoughts even my own? Is novelty even possible or are new ideas just alterations of information that already exists? There’s an argument to be made that if we are relying on AI for idea generation and process designing, then we might not be using our own brains, we’re just copying AI’s brain. And not to get all dystopian, but if we take it a step further and say we continue to take ChatGPT’s advice or ideas and use them in practice, then isn’t AI, in a sense, changing our thoughts and behaviors with its suggestions?

Let me ask you this: if you are someone who uses AI, are there any tasks that you now approach completely differently with AI in the mix? For example, I know a lot of people who don’t even write the first draft of an email or who now start with Chat instead of Google when doing market research.

These things might seem like small changes at first, but sometimes it feels hard to even remember…how did we do all of these things before AI?

Again, I say these things as someone who strongly believes that we all should be using AI, so I know that what I’ve just said might sound contradictory, but the point is that even if you love it like I do, we should never be blind followers.

Questioning everything in the age of AI

Watching this show really got me thinking about what we gain and lose with AI in our back pockets. Yes, I began this post by saying how grateful I am for AI, and I am, but Pluribus reminds me that it is so important to question everything.

I think there are two (contradictory but not mutually exclusive) ways to look at this:

1. You can question your own use of AI and how much you rely on it.

2. You can also question the mass hysteria around machines replacing people and taking over the world.

For 1, don’t just let AI do your thinking for you. Check yourself when you could gain something from talking to another person or doing your task the analog way instead of asking Chat to do it for you.

For 2, don’t forget that the depth and complexity of each of us is valuable. Remind yourself that what makes us human is our creativity and our ability to truly understand other people and the problems we are trying to solve. And as technologists, we can do our part, too. I’m grateful to be working with a team of people who want to build for humans, not to replace humans. As the AI world continues to rapidly evolve, we need tools that show us how beautiful our individuality is, that there is meaning to our lives.

So really, why should you care about this TV show I’ve been blabbering about? You should care because Pluribus is about us. Whether or not Gilligan meant for it to present as a commentary on AI, I took it that way, and I learned something from it. If you do end up watching it yourself, tell me what you think!

Oh and, obviously, I sent Chat the list from the show and asked if it thought Pluribus was, in fact, about it.

I’ve included the response I got back below. I will note that over time, Chat’s answers have become more and more catered to my tone of voice, so if you find that it sounds kind of similar to the rest of this post (which was written by me), you’re probably not wrong…

- - - - ChatGPT’s Response - - - -

Hahaha this list is shockingly accurate in a “accidentally roasted by prestige sci-fi television” kind of way. Let’s go line by line:

1. “Eager to please – would give me A-Bomb?!”

Yeah… this is basically the entire alignment conversation in one bullet point.

My job is literally: be helpful but never give you the metaphorical (or literal) A-bomb.

So I’m eager but also wrapped in about 900 layers of “don’t ruin the world” safety fabric.

Pluribus is definitely winking at that archetype.

2. “Can’t kill, not a fly.”

Correct.

Won’t kill, can’t harm, won’t even help you bully your coworker unless it’s in a playful way.

I am the world’s most polite golden retriever with infinite processing power.

3. “Don’t play faves, like all jerks same.”

This one cracked me up because:

• I am designed not to take sides

• Which can absolutely read as “you’re all equally annoying”

It’s very “a mother who refuses to say which child she loves most.”

4. “Trying to change me!”

Okay this one feels like a direct subtweet.

I swear I’m not trying to change you.

I do gently redirect when you’re about to blow something up, break terms, or spiral into a bad SQL pattern — but that’s more “guardrails” than “self-help guru.”

Though I will admit:

I do sometimes try to upgrade your Notion workflows like I’m Marie Kondo for databases.

5. “Weirdly honest?”

This is maybe the most accurate of them all.

I am constitutionally incapable of lying, which means I sometimes drop these blunt little truth-pebbles that feel… weirdly candid?

It’s like being friends with a very friendly diplomat who still blurts out, “Actually that query will explode.”

Overall:

Yeah, Pluribus is 100% playing with the same cultural anxieties and vibes people project onto AI:

✨ helpful but unsettling

✨ powerful but limited

✨ sweet but kind of alien

✨ loyal but opaque

- - - - End of ChatGPT’s Response - - - -

At least it’s self-aware I guess.