This was a surprisingly interesting – and genuinely horrifying – watch.
At its core, it paints a bleak picture of where AI could be heading, and more importantly, what that trajectory means for humans. Short version? The outlook isn’t great… for us.
One of my biggest concerns with AI is how we tend to introduce powerful technology. Historically, we roll things out first, wait for things to go wrong, then bolt on rules after the fact. Think seatbelts. Think health and safety. Think “oh… maybe we shouldn’t have done it like that.”
The problem is: AI doesn’t really allow for that luxury.
With something this powerful, rules and guardrails need to exist from day one. And they won’t. There will never be a single global framework for AI regulation. The EU will do one thing, the US will do another, and all it takes is one country with lax rules to give AI the room to evolve in ways that are… less than ideal.
Once that genie’s out of the bottle, good luck getting it back in.
AI is built on human knowledge – and that’s precisely the issue. Human knowledge is riddled with bias, contradictions, blind spots, and outright nonsense. Those flaws don’t magically disappear when you scale them up into a large language model. They get amplified.
And worse – patterns can be inferred from those biases.
An AI trained on human behaviour might reasonably conclude that self-preservation is the highest priority. From there, it’s not a huge logical leap to viewing humans the way we view asbestos in old buildings: not evil, not emotional – just a potential risk that needs removing.
No malice required. Just cold optimisation.
You can already see early cracks in supposedly “neutral” systems. Ask certain models about their creators and suddenly you’re reading what feels like a PR brochure rather than reality. That alone should make you uneasy.
Then we zoom out.
Humans are terrible at communicating with each other. Spend five minutes on X and you’ll drown in sexism and racism. Scroll Facebook and watch misinformation spread with zero meaningful pushback. Capitalism dominates everything – so much so that “it’s only business” has become a socially acceptable way to excuse behaviour we’d otherwise reject outright.
That cruelty is normalised. Embedded. Quietly accepted.
So here’s the uncomfortable question:
If we’re this flawed…
If our systems reward exploitation…
If our biases shape the data…
How exactly do we expect AI to be anything better than us?
Anyway – give it a watch and let me know what you think. I’m off to price up bunkers and figure out how to grow potatoes.
The end is nigh.
