#4 AI Expert Audit; The Time I made Notebook LM theorize about Five Nights of Freddy’s
Posted: February 4, 2026 Filed under: Uncategorized Leave a comment »The Source Material (I kind of went too far here):
we solved fnaf and we’re Not Kidding
https://www.reddit.com/r/GAMETHEORY/
How Scott Cawthon Ruind the FNAF Lore
https://freddy-fazbears-pizza.fandom.com/wiki/Five_Nights_at_Freddy%27s_Wiki
https://www.reddit.com/r/fivenightsatfreddys/
GT Live react to we solved fnaf and we’re Not Kidding
My Source Material, Why Did I Choose This?:
I actually chose materials that weren’t important to me, but they were when I was younger. I love listening to video essays and theories on various media. Whenever I was animating or doing a mundane art task in my undergrad, I would have that genre of video in the background to take a break from listening to the news (real important shit). It’s super silly stuff, but when I was a teenager, Game Theory first started getting BIG; seeing a huge channel discussing my favorite IPs, subverting and contextualizing their narratives felt very important. It really validated my feelings that video games were art.
However I am now grown, and I care far less about Five Nights of Freddy’s, now it feels like fun junk food for my brain. (Although teens and kiddos still care about the spooky animatronics, so it’s been a clutch move when bonding with the youths when I was a nanny.) I also hate AI, I hate it. I don’t hate automation; it makes life way better when done currently. I don’t think “AI” is done correctly; it’s mostly bullshit even down to the name. It’s a marketing strategy giving excuses to companies to fire workers and build giant databases that poison the land. I did not want to give Notebook LM anything “meaningful”. I didn’t want to let it in on the worlds I care about on my own volition. So, I gave it the silly spooky bear game that I know way too much about.
The AI Generated Materials
“Create an info graph of the official Five Night of Freddy’s Timeline with the information presented. Creating branches of diverging thought alongside widely agreed upon information.”
“Form a debate on what Timeline is the canon one FNAF.
Each host has to make their own original timeline.
Both hosts should sound like charismatic youtubers with dedicated channels to the video game and it’s lore.
Both Youtubers should use the words often associated with the Fandom and culture of FNAF.
Both hosts you have distinct personalities and opinions from one another.
Both hosts will have different opinions on whether the books should be used in lore making.”
1. Accuracy Check
What did the AI get right?
The basics. It was able to categorize the general hot topics (e.g., MCI or the Missing Child Incident, The Bite of 83’ and 87’, The Aftons…). It sometimes would match what theory goes with what Youtuber. It’s pretty efficient in barfing out information in bullet point fashion. I
What did it get wrong, oversimplify, or miss entirely?
The transcripts from the videos aren’t great; they don’t separate who is saying what, so when trying to describe the multiple popular theories out there and how they conflict, it struggles. When I had it made an audio debate where two personalities choose a stance to argue about from the materials I provided. It was pretty much mincemeat. Yes, both were referencing actual game elements but in ways to make no sense to the actual theories provided, the “hosts” argued about points no real person would argue about. In the prompt, I instructed one personality to use the books as reference while the other did not, and it took that and made 70% of the podcast arguing about the books. The mind map struggles to clarify what theory is and what is a canon fact. The info graph was illegible.
Were there any subtle distortions or misrepresentations that a non-expert might not catch?
Going back to the mind map, and in other words it doesn’t cite its sources well. It does provide the transcript it referred to, but the transcripts aren’t very useful as described above. It flips flopped between stated what as a theory and what was canon to the game (confirmed by the creators). If someone were to read it without much knowledge, they would be bombarded with information that conflicts, isn’t organized narratively, and stated in context of its origin.
2. Usefulness for Learning
If you were encountering this material for the first time, would these AI-generated resources help you understand it?
Semi-informative but not at all engaging.
What do the podcast, mind map, and infographic each do well (or poorly) as learning tools?
Both podcast and mind map were at least comprehensible; the info graph was not.
Which format was most/least effective? Why?
The podcast is the most effective; there was some generated personality to distinguish the motivation behind certain theories, not great distinctions but more than nothing.
3. The Aesthetic of AI
It’s safe to say Youtubers and podcasters are still safe job wise. Hearing theories about haunted animatronics in the format and aesthetics of an NPR podcast was deeply embarrassing. Hearing a generated voice call me a “Fazz-head” was demoralizing to say the least.
They made pretty bad debaters too. The one who was presumably assigned the role of “I will only use the games as references” at one point waved away their opponent’s claim with the response, “yeah but that’s if you seriously take a mini game from 10 years ago”.
It took out all of the fun; there were no longer cheeky remarks of self-depreciating jokes about the silliness of the topic and efforts. Often theorists will acknowledge Scott Cawthon did not think these implications fully out, that this effort may be rooted in retcons and wishful thinking, but it’s still fun. The hosts and mind map acted like they were categorizing religious text, and it was remarkably unenjoyable to sit through.
4. Trust & Limitations
AI is good at taking (proven) information and organizing it in a way that is nice to look at. It’s great for schedules or breakdowns. It sucks at just about everything else. I only really have benefitted from AI when it comes to programming; it’s really nice to have an answer to what is wrong with your code (even if it’s not always right; it usually leads you past the point of being stumped).
When it comes to art, interpretation, and comprehension, I wouldn’t recommend AI to anyone. If you are making a quiz, make it yourself. The act of making a quiz based off study topics will increase your comprehension far more than memorizing questions barfed out to you. If you don’t have the time to produce something, then produce something you can with the time you have or collaborate with someone who can produce with you. Use AI to fix your grammar (language or code), use AI to make a schedule if you suffer from task paralysis, but aside from accommodations and quick questions, leave it alone.