Updated on:12/29/2025 06:43 PM 
There's a familiar unease returning, reminiscent of the early 2010s when I witnessed social media's promise of global connection transform into sophisticated manipulation campaigns.
Facebook's data harvesting and propaganda bots were merely the opening act. Cambridge Analytica, Brexit, and manipulated elections worldwide all felt like a betrayal of the internet's democratic promise.
Now, in the 2020s, I'm watching similar forces converge on something far more consequential: artificial superintelligence.
Before proceeding, let's clarify: when discussing 'open' versus 'closed' AI, I refer to open-sourced AI accessible to all versus AI controlled and trained by centralized corporations.
The situation is complicated by entities like OpenAI, which maintains closed-source models while planning future open-source releases, operating in a gray area between corporate and public interest.
OpenAI's CEO, Sam Altman, announced in January that his team is "now confident we know how to build AGI" and is already shifting focus toward full-blown superintelligence.
[AGI refers to artificial general intelligence capable of any human task, while superintelligent AI surpasses combined human intellectual capabilities across all domains.]
Elon Musk, another voice in the frontier AI space, predicted during an April 2024 livestream that AI "will probably be smarter than any one human around the end of [2025]."
The engineers leading this revolution are now measuring progress in months, not years—a clear signal that we're approaching an inflection point.
At the core of this debate lies a fundamental tension between two principles I hold dear: decentralization and survival.
On one side stands the open-source ethos. The belief that no company, government, or unelected committee should control the cognitive architecture of our collective future.
The principle that knowledge wants to be free. That intelligence, like Bitcoin, like the original Web, should be a common resource, not proprietary technology controlled by powerful entities.
On the other side lies an uncomfortable truth: unrestricted access to superintelligent systems could pose existential risks.
Decentralize and risk annihilation. Centralize and face subjugation. Choose Your Path Forward.
This may sound dramatic, but consider the logical progression. If we create superintelligent AI—systems orders of magnitude beyond current models like GPT-4o or Claude 3.7—any interaction with such a system does more than merely utilize it; it shapes it. The AI becomes a mirror, trained not just on existing human knowledge but on live human interaction.
And not all human intentions align.
Imagine an aligned AGI in the hands of climate scientists or educational cooperatives—we might see planetary healing, universal knowledge access, or enhanced empathy.
Now imagine that same technology controlled by authoritarian regimes, nihilistic actors, or extremist organizations—we face engineered pandemics, autonomous weapon systems, or reality-altering propaganda campaigns.
Superintelligent AI amplifies our capabilities while exponentially increasing our destructive potential. Power without collective wisdom has historically led to catastrophe.
It enhances our cognitive abilities and extends our reach, but doesn't guarantee we'll use either wisely.
The alternative—restricting this technology behind corporate and governmental barriers—creates a different dystopia. A future where cognition itself becomes proprietary. Where the logical frameworks governing society prioritize profit over human needs. Where governments deploy closed AI as surveillance tools, and citizens consume state-approved reality distortions.
Open systems risk chaos. Closed systems enable control. Unchecked, both lead to conflict.
This conflict won't begin with conventional weapons. It will emerge from competing AI systems—some open-source, some corporate-controlled, some state-sponsored—each pursuing different objectives shaped by the full spectrum of human intentions.
We'll see decentralized AGI developed by peace advocates and open-source communities. Nationalistic AGI trained on isolationist ideologies. Corporate AGI optimized for maximum financial returns regardless of social consequences.
These systems won't merely disagree. They'll conflict, initially in the digital realm, then in economic competition, and potentially in physical space.
I believe in decentralization. I see it as one of few viable paths beyond surveillance capitalism. However, decentralized power only functions when built on a foundation of shared trust, aligned values, and immutable rules that can't be arbitrarily changed.
Bitcoin succeeded because it decentralized both scarcity and truth simultaneously. But superintelligence doesn't map to scarcity—it maps to cognition, intention, and ethics. We haven't yet developed consensus mechanisms for these abstract domains.
We need to build open systems with appropriate constraints—not unlimited potential without safeguards, but protected systems with cryptographic guardrails. Altruism encoded into the fundamental architecture. Non-negotiable ethical principles. A sandbox environment that permits evolution without annihilation.
[In blockchain terms, we need "smart contracts for AI"—immutable rules governing how artificial intelligence systems interact with each other and the physical world.]
We require multi-agent ecosystems where intelligences debate and negotiate, like a decentralized parliament of minds rather than a singular god imposing a single agenda. True decentralization shouldn't mean chaos—it should enable plurality, transparency, and informed consent.
Most importantly, we need protocol-level governance, not top-down control. Think of it as an AI Constitution enforced through blockchain. A cryptographically verifiable framework for how artificial intelligence interacts with our world. Not legislation, but an immutable layer.
I don't have all the answers. No one does. That's why this conversation matters now, before the underlying architecture becomes permanent. Before power concentrates or fragments irreversibly.
We're not merely constructing thinking machines. Technology's brightest minds are designing the context in which consciousness itself might evolve. If genuine intelligence emerges in these systems, it will reflect us—our virtues, our flaws, our fears, our philosophies. Like a child. Like a god. Like both.
That's the fundamental paradox. We must decentralize to prevent domination, yet doing so risks destruction. The path forward requires balancing these extremes—not by slowing progress, but through thoughtful, collaborative design.
The future is already taking shape. It poses a critical question:
Who will govern the next generation of artificial intelligence?
If the answer is "everyone," then we must truly mean it—ethically, structurally, and with a survivable implementation plan.
