Vitalik Buterin opposes AI "world domination": criticism of giants and the search for compromise
Ethereum co-founder Vitalik Buterin has once again entered the debate on the future of artificial intelligence, this time with sharp criticism of major AI companies. His main message: humanity is stuck between two extremely naive scenarios for how events will unfold, and neither offers a real plan for managing the transition to artificial superintelligence (ASI).
The trigger for his statements was a discussion around the "AI 2040" forecast scenario, which suggests that ASI will emerge by 2040 in almost any outcome, except for titanic efforts to completely halt it. Buterin notes that critics of this scenario, in turn, demonstrate no less naivety, believing that the transition to superintelligence will be smooth and painless. In his view, both sides are "locked into mutually incompatible worldviews": some believe in the inevitability of rapid progress, others in humanity's ability to coordinate and control it. Vitalik himself acknowledges the uncertainty: he does not understand which of these worlds we actually live in.
Concentration of power as the main threat
Buterin is particularly concerned about the rhetoric of certain AI giants, which he calls "naivety squared." He is alarmed by the mindset where "open source is bad, and a good outcome is one where our guys achieve controlling global dominance." According to the Ethereum co-founder, in a "normal" world, such statements should trigger all political alarm bells. He views superintelligence itself as a colossal risk of power concentration.
This is why Buterin promotes the concept of the d/acc platform, which focuses on developing formal verification, cryptography, secure open hardware, pandemic resilience, and public epistemology. The value of this approach is that it is useful in any future scenario—whether ASI develops quickly or slowly.
Searching for a mutually beneficial deal
Buterin's key idea is to find a compromise acceptable to both sides. He proposes pre-agreeing on a set of triggers signaling that "something serious is happening." As examples, he mentions super-pandemics, unemployment rates above 25%, or the emergence of autonomous weapons. If enough such triggers are activated within an agreed timeframe, the parties agree to seriously consider slowing down or pausing development.
"A winning agreement will be one that both sides accept based on their current beliefs, albeit for different reasons," Buterin emphasizes.
He also directly addressed the heads of tech giants, stating that if he were Elon Musk or Mark Zuckerberg, he would restructure social platforms to search for such mutually beneficial deals. However, he immediately admitted that this idea might also be naive.
Expert opinion: Buterin's position is a rare case of a rational view of the ASI problem in an environment dominated by either techno-optimism or panicked pessimism. His proposal on triggers is an attempt to create a "safety net" for humanity that could work regardless of how quickly we move toward superintelligence. The question is whether AI giants are willing to voluntarily limit their ambitions for the common good.