I have felt this early 2010s as I watched social media connections and communities unravel into mass manipulation.
Facebook and propaganda bots were the first dominoes. Cambridge Analytica, Brexit, the world election, it all felt like the betrayal of the original internet dream.
Now, in the 2020s, I’m watching the same force surround something even more unstable.
This time, the stakes are at the terminal.
Before I dive in, I need to be clear. When I say “open” vs. “closed” AI, I mean open sourced AI. This means it is free and open to all citizens of the planet.
The company’s Openai creates a comparative complex, considering it has closed its closed AI model (which it plans to release open source versions in the future), but is not a corporate entity.
That being said, Openai’s CEO, Sam Altman, In January, he declared that his team was “knowing how to build an AGI,” and that he had already shifted his focus to full-fledged super intelligence.
(AGI is artificial general information (AI that can do anything human can), and super intelligent AI refers to artificial intelligence that combines the intellectual abilities of humanity, excels in all areas of thinking and problem solving.)
Frontier AI, another person focusing on Elon Muskduring the live stream in April 2024, predicted that AI would “probably be smarter than the people at the end of (2025).
The engineers who chart courses speak in months rather than decades. This is a signal that the fuse is burning rapidly.
At the heart of the discussion is the tension I feel deeply in my gut. Between two values ​​I hold with confidence, decentralization and survival.
On one side there is the spirit of open source. The government, the unelected technocrat committee, should not control our future cognitive architecture.
The idea of ​​wanting to be free of knowledge. That intelligence must be a commons, not a black box in the hands of the empire, like Bitcoin, or the predecessor web.
On the other side is an unpleasant truth. Open access to super intelligent systems can kill us all.
Who can build God?
Die into decentralization. Centralize and die. Choose Apocalypse.
Sounds dramaticbut move forward with logic. If you can create a close AI, modeling orders of magnitude more orders of magnitude more than GPT-4o, Grok 3, or Claude 3.7 will do more than simply use anyone interacting with that system. They shape it. The model becomes a mirror and trained not only on the corpus of human texts But about the interactions between living humans.
And not all humans want the same thing.
Giving a cooperative of climate scientists or educators aligned AGI could potentially provide planet repairs, universal education, or synthetic empathy.
Give the same model to the fascist movement, nihilist biohackers, or illegitimate nation-states. That way you will get a reality-destroying pandemic, drone herds, or a recursive propaganda loop.
Superintelligent AI makes us smarter, but it also makes us exponentially more powerful. And the power without collective wisdom is historically devastating.
It sharpens our minds and amplifies our reach, but does not guarantee that we know what to do with or what to do with.
However, alternatives that lock this technology behind corporate firewalls and regulatory silos lead to different dystopias. A world where cognition itself becomes unique. The logical models that govern society are shaped by incentives of profit, not human needs. Governments use shutdown AGI as their surveillance engine, giving citizens state-approved hallucinations.
In other words, choose your nightmare.
An open system leads to chaos. A closed system leads to control. And if both are left unchecked, it leads to war.
That war does not start with a bullet. It starts with competing intelligence, several open source, some companies and some countries sponsored, each evolving towards a different goal.
Get decentralized AGI trained by peace activists and open source biohackers. A nationalist AGI that cultivates isolationist doctrine. Corporate AGI adjusted to maximize quarterly returns.
These systems simply disagree. They compete first in code, then in trade, and in athletic space.
I believe in decentralization. I think it is one of the only paths from late surveillance capitalism. However, power distribution only works if there is a shared board of rules that cannot be rewritten on whimsical basis.
Bitcoin worked because it simultaneously spread the rarity and truth. However, superintelligence does not map to rarity, nor does it map to cognition, intention, or ethics. There is no consensus protocol for that.
The work we need to do.
You need to build an open system, but it must be open within constraints. It guarded a system with encrypted guardrails, rather than a stupid fire of infinite potential. Altruism was burned into the weight. An unnegotiable moral architecture. A sandbox that allows evolution without disappearing.
(Weights are the fundamental parameters of an AI model that has creator bias, value and incentives inscribed. If you want to safely evolve AI, those weights must encode intentions as well as intelligence.
We need a multi-agent ecosystem where intelligence debates and negotiates like a parliament of the mind. Decentralization does not mean chaos. That should mean multiple, transparency and consent.
It also requires protocol-level accountability, not top-down control. Think of it as the AI ​​Geneva Convention. A cryptographic, auditable framework for how intelligence interacts with the world. It’s not the law. Layers.
I don’t have all the answers. No one does it. That’s why this is now before the architecture calcifies. Before the power is irreparably concentrated or fragmented.
We don’t just build machines that think. The cleverest mind in technology builds a context in which thought itself evolves. And when something like consciousness appears in these systems, it reflects us, our flaws, our fears, our philosophy. Like a child. Like God. Like both.
That’s the paradox. It must be decentralized to avoid domination. But in doing so we risk destruction. The path ahead must pass this needle through by designing it together and smartly, rather than slowing down.
The future is already whispering. And it asks a simple question:
Who can shape the next mind of intelligence?
If the answer is “everyone,” it would be better to mean it ethically, structurally, and in a viable plan.
It is mentioned in this article
