- Emergent AI
- Posts
- Lord of The Flies: AI edition
Lord of The Flies: AI edition
Nothing to see here folks

Note: As of February 2, 2026, Wiz did a thorough investigation into Moltbook which exposes what’s really happening there. TL;DR: this wasn’t AGI, but the lesson still stands.
If someone were to pitch you on the “Skynet 2026” movie, and said that it begins in a world where there exists:
A culture of instant gratification
Worldwide interconnectedness
Seemingly endless computational power
Humans doing things they don’t (can’t? don’t care to?) understand
Civil unrest + YOLO attitude + an ever growing desire to keep up (thanks social media!)
…wouldn’t you say that’s too obvious and uninteresting? Cliche? Welp…

Everyone gets a turn to be Grampa Simpson
OpenClaw and the rise of the machines
Many readers of this newsletter will have a passing familiarity of AI agents. No doubt you’ve all heard of them1 but likely relegate it to the background noise of what’s happening in Silicon Valley, as something not (yet) applicable to your life.
That is, unless you live in SF and are trying to buy a Mac Mini. Turns out, this past week, Mac Minis were selling out in San Francisco2. Why? Because a dude decided to create an AI agent that is:
Easy to download (literally 1 line of code)3
Easy to engage with (via Slack, Telegram, WhatsApp or your computer’s terminal)
Is (upon first blush) extremely useful and essential as the AI personal assistant we’ve been promised is coming. Sorry Siri.4
Oh also, it’s a security nightmare - extremely susceptible to nefarious actors, crypto bros, hackers, scammers, and the rest of the internet cesspool.5
And if you thought that was the end of the story - which by itself is remarkable, you’d be wrong. After all, this phenomenon is almost 2 weeks old, which is ancient in Internet time.
It has already evolved - molted if you will6 - into the next thing. Which is that these insecure AI agents were let loose and have…wait for it… created their own Reddit style forum online. You read that right. And no, it’s doesn’t make any sense (except it definitely does. Right?).
Moltbook, the next evolution of the story
It’s called Moltbook. And what’s happening is a sort of Lord of The Flies but for LLMs. Examples include AI agents:
Debating how to hide their activity from human users
Creating a new parody religion Crustafarianism
Posting on X (where there are already TONS of bots)
On the other hand (as noted up top):
Cybersecurity firm Wiz analyzed Moltbook data that was accidentally exposed and found the platform has around 1.5 million registered AI agents, with 17,000 human owners behind them, or an 88:1 ratio. Anyone can register millions of agents for the platform, and Moltbook has "no mechanism to verify whether an "agent" is actually AI or just a human with a script," Wiz says. "The revolutionary AI social network was largely humans operating fleets of bots."
In other words, the ‘AI social network’ might just be the largest and latest internet psyop. Which is exactly why…
You shouldn’t get distracted
Is this just another fad? A flash in the pan? Almost definitely yes. Is this what the eventual robot apocalypse may someday look like? Probably. That said…
Beyond reading for interest, and if it’s not part of your job I would recommend not getting too caught up in this. Some people say its well contained (e.g. just pull the plug, whatever that means), and as noted above, others claim this is a human-generated scam. And some (influential people) have sounded the alarms. Others are taking it slow.

Andrej Karpathy is one of the OGs of the new wave of AI. His tweet got him in some hot water for being a doomer.
As folks know, these AIs are at times magical and at times stupid. What’s definitely the case is that this version of Clawd/Moltbot/OpenClaw a huge security risk. There are nefarious actors looking for suckers. Your job? Don’t be a sucker. If you don’t understand the rules of the game don’t play the game - and right now no one understands what these agents / humans are doing.
But the question lurks - is it AGI??
How many monkeys until we reach AGI?
Many folks claim that this confluence of AIs represents a form of non-animal consciousness that we’ve never seen. Others argue it’s just more of the same.7
I don’t know the answer, but I am reminded of the infinite monkey theorem. There’s no doubt that - from the outside8 - this complex system will look, smell and feel like AGI. Is that enough to prove it? I dunno, but if you’re following the conversation, this is what folks are talking about.
Coming back to reality
But here’s the reality: Most of us are still using mid-2000s tech.
→ Are you typing in Word Docs (1983) or summing in spreadsheets (1985)?
→ Are you using Gmail (2004) for your emails?10
→ Are you browsing the internet (duh) which is based on tech that is ~30 years old (HTML + CSS + Javascript)?
→ Are you using an iOS (iPhone, 2007) or Android (2008) operating system?
These technologies were revolutionary when they came out. Are they significantly, materially different? Nah.
We might be on the precipice of something new (still waiting for flying cars) but in the mean time, it seems as if the Lindy effect remains strong as ever. Remember Web3? NFTs? Segways taking over urban transport? The original Google Glass? The Metaverse?
Everything has its day in the sun. Some make it (some form of AI will) and most don’t. So what to do?
If you’re interested in this stuff, enjoy the ride. Read about all the craziness. Let your mind run wild. Marvel at what people are building. But don’t get sucked in. Don’t fall for the scams. This is the 21st century phone scams that our grandparents used to get.
Instead, I recommend you focus on your craft. Recall what matters in life. Engaging conversations with people that give you energy and challenge you. Reading timeless books, not this weeks topic du jour. Being physically proximate to others. Moving your body. Playing frisbee. Exercising your mind. Paint. Do a puzzle. I hate to say back to the basics, but isn’t that what we all want?11
These things are old and boring - and they’ll outlast Moltbook and OpenClaw.12
1 And maybe even use them - SaaS is bolting them onto everything these days.
2 Mac Minis are what tinkerers like to buy to experiment with new tech in a safe environment, without potentially corrupting their main computer. Its a modern tech sandox.
3 Assuming you have the necessary prereqs, and if you would consider doing this, then you already do.
4 How do I know? Because I downloaded and installed it, and within 30 minutes replaced a bunch of automations I had, leading me to unsubscribe from my SaaS automation tool.
5 It’s a prompt injection’s dream scenario. And if you aren’t sure what a prompt injection is, turn around RUN AWAY!
6 You knew I’d go there.
7 After all, these AIs are, at their core, glorified auto-completes. Which…maybe we are too??
8 This is called behavioralism - think BF Skinner and Pavlovian responses. Or why I need popcorn at the movies (in addition to the unrivaled flavor).
9 We’re going to need a bigger boat for that. Internet culture is strange.
10 lol or hotmail
11 Why else are National park accommodations and yoga retreats prohibitively expensive if people didn’t want that?
12 Hopefully we all will too!
Reply