TLDR:Buterin says power concentration threatens humanity more than AI intelligence itself.He backs open-source mandates to prevent AI labs from dominating development control.Buterin supports d/acc, prioritizing cryptography, secure hardware, and pandemic resistance.He proposes pre-agreed triggers to resolve AI safety and acceleration disagreements.Vitalik Buterin warned that the biggest risk tied to artificial intelligence may not be the danger most people focus on today.The Ethereum co-founder argued that concentrated power, not runaway machine intelligence alone, poses the deeper threat. Buterin’s comments came as debate intensified around AI 2040 predictions and superintelligence timelines. He suggested public discourse often overlooks how control over AI systems could reshape global power structures entirely.Buterin Identifies Power Concentration As The Overlooked AI ThreatButerin explained that most discussions center on whether superintelligence itself will prove dangerous. He said fewer people examine what happens if a small group controls that intelligence. According to Buterin, humanity’s collective bargaining power could drop to zero under certain conditions. This occurs if AI systems can outperform humans across every conceivable task.One thing I find striking in the discourse between AI 2040 and its detractors is that the two seem to be locked in to totally incompatible worldviews of how fast and how much of a big deal AI progress is:* In AI 2040, every scenario sees superintelligence of some kind emerging…— vitalik.eth (@VitalikButerin) July 11, 2026The Ethereum co-founder criticized narratives suggesting dominance by a few AI labs represents an acceptable outcome. He compared this framing to political warning signs that would raise concern in any other setting. Buterin said assuming a smooth transition to advanced AI carries its own naivety. He argued this assumption receives far less scrutiny than proposals for slowing AI development.Buterin pointed to specific mechanisms within AI 2040 that address this power imbalance directly. He referenced updated provisions mandating open-source access to AI development. This shift reduces the chance that a handful of companies control transformative technology. Buterin called this revision a meaningful improvement over earlier centralized frameworks.He also cited “mutually assured compute destruction” as a mechanism worth serious consideration. This concept would allow multiple actors to trigger a global compute slowdown collectively. Buterin said this differs sharply from systems letting select groups disenfranchise others selectively. He framed this distinction as central to evaluating any AI governance proposal fairly.Ethereum Founder Backs Decentralized Approach To Manage AI UncertaintyButerin described his support for d/acc, a decentralized defensive strategy against AI risks. This framework prioritizes formal verification, cryptography, and secure open hardware development. It also includes pandemic resistance, food security, and public epistemics as core components. Buterin said these measures hold value regardless of how AI timelines actually unfold.He proposed establishing pre-agreed triggers to resolve disagreements between AI accelerationists and cautious observers. Examples included thresholds like severe pandemics or unemployment surpassing twenty-five percent. Once triggers occur or fail to materialize, legitimacy would shift toward whichever side predicted accurately. Buterin suggested social media platforms could host these negotiations directly.According to Buterin, this method would let more people participate beyond governments and executives. He argued current AI governance debates exclude too many stakeholders from meaningful input. Buterin acknowledged his own proposal contains elements that critics might label unrealistic. Still, he expressed sympathy toward efforts attempting to address AI risks responsibly and transparently.Buterin closed by noting no existing plan fully escapes accusations of naivety. He said humanity faces a choice between imperfect frameworks rather than a perfect solution. Buterin indicated he remains open to supporting proposals that reduce concentrated control over AI systems.The post Vitalik Buterin: AI’s True Risk Lies in Who Controls It, Not Its Intelligence appeared first on Blockonomi.