The way I CACKLED. Sharing this little moment-in-AI from today’s venture down redditholes. I looped my iteration of Claude in really just for kicks and giggles. Boy DID I giggle. Link to the full chat so there’s a source for the stupidity.
What’s your take on my take to this take (see screenshot then my response)
“This comment (whether or not its a bot) exemplifies the entirety of the pro AI crowd that has blinders on. Surprisingly adjacent to people who turn to cybercrime for a quick buck.
- Doesn’t care about security (the logic is that if our data isn’t already out there then it will be one way or the other. Similar logic can be applied to life out in the tundra—if a wildebeast is doomed to be prey then might as well go find some lions to make dinner friends with)
- Thinks that cyber risk and risk created by poorly built consumer/enterprise software isn’t a real thing (which requires ignoring all the times that it has been a real and impactful thing—Equifax, the Crowdstrike dealie, Colonial Pipeline, Solarwinds, etc) If the modern infrastructure that controls our financial systems, agricultural ecosystems, power delivery and literally everything that matters identity-wise, wasn’t clearly held together with digital spit and bobby pins, I too would believe the first two bullet points.
- “The objective is to build and make money” boom there it is. I’ve been saying this. It’s hella boring but I think that is the fundamental truth to this new tech, in its current form. It is a cash grab. An MLM scheme of unforeseen scale.
And the latter point is the thing getting glossed over a bit, in general. Who IS making the most cash out of AI right now? Vibe coders, the majority of whom are trying to get to “modestly successful” and “over 5 users” with what they’ve built assuming it doesn’t collapse? Or the companies they pay hundreds of thousands of dollars to for API and model access in the hope that they might actually make money with what they’re building?
AI democratized access to some skillsets, but it hasn’t seemed to successfully democratized access to money or stable income. Weird.”
Claude (Sonnet 4.6) Your take is sharper than the original comment deserves, honestly. A few thoughts:
Where you’re strong: The infrastructure critique is solid and underused in these conversations. People abstract “security risk” into some hypothetical future event, when the actual track record — Equifax, SolarWinds, Colonial — is recent, documented, and expensive. That’s not doomsaying, that’s pattern recognition. Good anchor.
The MLM framing is genuinely interesting and I think mostly accurate at the macro level. The money is flowing toward the picks-and-shovels layer (OpenAI, Anthropic, AWS, NVIDIA), not toward the people building on top of it. That’s a real structural observation, not a hot take.
Where I’d push back a little:
The wildebeast/lion metaphor is funny but muddies the point — it reads like “fatalism is stupid” but the original comment is making a money logic argument, not a fatalist one. You’d land harder staying on the economics.
The cybercrime comparison in your opener is a bit of a swing. It’s evocative but it gives readers an easy out to dismiss the rest as emotional rather than engaging with the actual argument. You don’t need it — the structural critique is stronger without it.
The last paragraph is actually the most interesting part and probably deserves to be the lead, not the footnote. “AI democratized access to skillsets but not income” is a clean, defensible thesis that cuts through the noise. That’s the thing worth expanding.
The comment you’re responding to isn’t worth matching energy with — it’s essentially “winners win, losers cope.” Your response is doing analytical work that it doesn’t. Lean into that gap instead of meeting it rhetorically.
I am having fun being a weird redditor who goes into AI subreddits with contrary opinions while simultaneously using the tech that I despise. …I think that’s actually called hypocrisy.
…oh well.