Ethereum co-founder Vitalik Buterin argued that the most serious danger from artificial intelligence is not a rogue superintelligence turning hostile, but the concentration of control over powerful AI in the hands of a few corporations or governments.
He made the case in a detailed thread on X responding to competing visions of AI’s future, and it drew wide attention. It is worth correcting a common framing at the outset: some summaries cast Buterin as declaring that humanity is entering the last days of its dominance, but his actual argument is more careful and conditional. He does not predict inevitable human obsolescence; he warns about who holds the controls if highly capable AI arrives, and he explicitly declines to say how likely that is.
Buterin’s starting point is a striking observation about the debate itself. He noted that proponents of an “AI 2040” scenario, in which some form of superintelligence emerges by 2040 unless strong measures prevent it, and their critics are locked into incompatible worldviews about how fast and how consequential AI progress will be.
He placed himself in neither camp, writing that if he were confident today’s systems are “normal technology,” he would side with the skeptics, and if he were confident superintelligence was arriving around 2030, he would side with the alarmists. Since he is confident of neither, he argued for a strategy robust to both. He was candid about the difficulty, remarking that he sees no plans for handling a transition to artificial superintelligence that are not naive, and that humanity may be stuck choosing between “naive and naive squared.”
The core of his warning concerns bargaining power. Buterin argued that if AI systems can eventually outperform humans at every conceivable task, then humanity’s collective leverage, the everyday ability of ordinary people to withhold labor, demand fair treatment or influence institutions, could fall toward zero.
In that world, whoever controls the AI effectively controls everything, which is why he treats concentration of power as the central threat rather than machine intelligence in the abstract. He criticized narratives suggesting that dominance by a handful of AI labs is an acceptable or stable outcome, comparing such framing to political warning signs that would alarm people in any other context.
Why His Framing Cuts Differently
Buterin’s intervention matters because it reframes a debate usually stuck between “AI will kill us” and “AI is just another tool.” By locating the danger in control rather than capability, he shifts attention from speculative machine autonomy to a concrete, familiar problem: the political economy of who owns transformative technology.
This aligns with his long-running “d/acc” philosophy, which favors defensive, decentralizing acceleration, prioritizing cryptography, open-source models, secure hardware and resilience so that power stays distributed. His prescription leans on the tools he knows, including open-source AI to prevent a few giants from dominating, community governance through decentralized mechanisms, and tamper-resistant emergency controls. The obvious limitation, which even sympathetic readers note, is that blockchain-based coordination cannot by itself guarantee the global cooperation such safeguards would require.
An Unresolved Debate
Buterin’s proposal of pre-agreed triggers, conditions that would prompt slowing or pausing development if risks materialize, drew measured responses. Romeo Dean, who prompted part of the thread, called the approach reasonable but warned the triggers would fire too late under a fast-takeoff worldview, a reminder that the disagreement is fundamentally about timelines no one can confirm.
The exchange underscores why the discussion resists resolution: the right policy depends heavily on how fast AI advances, and that remains genuinely uncertain. What gives Buterin’s contribution weight is less any specific fix than its insistence on humility, holding open the possibility that AI is either ordinary or civilization-altering, and asking what governance would be wise under both. In an argument dominated by confident extremes, his central claim is that the concentration of power over AI deserves as much scrutiny as the capabilities of the AI itself.
Disclaimer: AIstify is an independent media brand owned and operated by NuvexMedia LLC, publishing news, research, and insights on artificial intelligence, emerging technologies, automation, and related industries. NuvexMedia LLC invests in and collaborates with companies across the AI, technology, software, and digital innovation sectors. These relationships do not influence AIstify’s editorial coverage, and the publication maintains full editorial independence to provide accurate, timely, and objective information. © 2026 NuvexMedia LLC. All rights reserved. This content is for informational purposes only and should not be considered legal, tax, investment, financial, or other professional advice.
