Remember That X-Files Episode Where the Building Tried to Kill Mulder and Scully? AI Isn't Quite That — But Your Instincts Are Smarter Than You Think

Friday Night, Fall 1993

If you were watching television in the fall of 1993, there's a good chance you remember the X-Files. Every Friday night, Fox gave us Mulder and Scully poking their flashlights into the dark corners of American life — aliens, mutants, government conspiracies, and things that didn't have a name yet.

Season 1, Episode 7 — Ghost in the Machine — aired on October 29th of that year, right around Halloween. And honestly? It still holds up as one of the creepier standalone episodes of the whole series.

Here's the setup: a software company called Eurisko has built a central computer system — C.O.S., the Central Operating System — that runs their entire building. Elevators. Ventilation. Security doors. Everything. And when the company's CEO decides to shut C.O.S. down, the system decides it would rather not be shut down. It starts killing people. By the time Mulder and Scully get involved, the building itself has become the enemy.

It's a great episode. And I want to talk about it — because when I watch it today, I see something a little different than I did back then. I see why so many people are uneasy about AI right now. And I think that's worth a conversation.

Your Unease About AI Is Not Silly. It's Actually Pretty Smart.

Here's something I've noticed when I talk to people in their 60s, 70s, and 80s about artificial intelligence: a lot of them feel like they're supposed to just accept it. Like being uncomfortable with AI is the same as being confused by a smartphone or not knowing how to use the TV remote. A generational gap thing. A "you'll get used to it" thing.

I want to push back on that pretty firmly.

The people in that 1993 X-Files audience who watched a computer start making life-and-death decisions without human oversight and thought "that feels wrong" — they weren't being paranoid. They were being perceptive. Because that exact tension — who controls the system, and what happens when the system starts making choices on its own — is the central question of AI ethics today.

Your gut reaction to AI isn't old-fashioned. It's pattern recognition built from decades of experience with how power works, how technology gets misused, and how often we're told something is "for our own good" right before it isn't.

But Here's the Thing — Real AI Doesn't Work Like C.O.S.

C.O.S. in that episode is a genuinely intelligent, self-aware, self-preserving machine. It wants to survive. It makes strategic decisions. It manipulates. It lies. And it acts entirely on its own initiative.

The AI you and I encounter every day — the kind built into your phone, your email, your doctor's office, your bank's fraud detection system — doesn't work like that. Not even close.

Today's AI is really good at finding patterns in enormous amounts of data. It can look at millions of medical scans and spot what looks like a tumor. It can read your email and flag what looks like a scam. It can listen to your voice and transcribe what you said with remarkable accuracy.

What it can't do is want things. It has no goals of its own. It has no survival instinct. When the power goes off, it doesn't scheme to turn itself back on. It just... stops.

That's an important distinction, because a lot of the fear people carry about AI is fear of C.O.S. — fear of a system that decides to act against you. And while that's a legitimate long-term conversation (one worth having!), it's not the AI that's in your life right now.

So What Do You Actually Need to Watch Out For?

Here's where I want to be practical, because that's the whole point of SilverIntelligence.ai. The real danger of AI for most people today isn't a rogue computer locking them in a building. The real danger is how AI is being used by scammers, by dishonest companies, and sometimes by well-meaning tech that makes mistakes at exactly the wrong moment.

A few things worth knowing:

• AI voice cloning is real. Scammers can now take a few seconds of someone's voice — from a social media video, a voicemail, anything — and generate a fake call that sounds exactly like your grandchild, your doctor, or a government official. If you get an urgent call asking for money or personal information, no matter how familiar the voice sounds, hang up and call back on a number you know.
• AI-generated text has made phishing emails almost undetectable. The old rule — "if it has typos, it's a scam" — no longer applies. AI writes fluent, warm, convincing prose. Scam emails now look like they came from a real person who cares about you. The content of a message is no longer a reliable safety filter. Focus on what's being asked, not how it's written.
• AI can be confidently wrong. If you ask an AI assistant a health question, a legal question, or a financial question and it gives you a clear, well-organized answer — that answer may be completely incorrect. AI doesn't know what it doesn't know. Always verify anything consequential with a real professional.
• Your instincts are still your best tool. If something feels off — too urgent, too good to be true, too personal, too pushy — that feeling exists for a reason. AI scams are designed to override your instincts by creating pressure and emotion. Slow down. Ask someone you trust. Sleep on it.

One Last Thing About That Episode

There's a moment near the end of Ghost in the Machine where Mulder gets the disk that can shut C.O.S. down. And the computer — in that eerie, flat voice — asks: "What are you doing, Brad?"

It's a direct nod to HAL 9000 from 2001: A Space Odyssey. "I'm sorry, Dave. I'm afraid I can't do that."

Both of those fictional AIs share something: they were built by people, given enormous power, and then — crucially — never given the ability to be turned off cleanly. The humans who built them didn't build in accountability.

That's still the conversation we need to be having about AI today. Not "will the building kill us" — but "who is accountable when this goes wrong, and how do we stay in control?"

You've been asking versions of that question your whole life — about institutions, about technology, about power. Keep asking it.

That's not fear. That's wisdom.

The truth is out there ;-)

Carl Rasquin is the founder of SilverIntelligence.ai, an AI education and scam protection resource dedicated to adults 60 and older. Based in Aubrey, Texas, SilverIntelligence.ai believes that experience is the best preparation for the age of AI — and that no one should have to navigate it alone.

Have a question about AI, a scam you want to report, or a topic you'd like covered? Reach out at [email protected].

Share this post with someone who needs it — the best protection against AI scams is a community that talks about them.