Vitalik Buterin has made a new statement about the future of artificial intelligence. He suggested establishing in advance the conditions under which the creation of the most powerful AI systems could be slowed down. According to him, agreeing on these rules ahead of time would help avoid disputes if AI development accelerates.
What Measures Is Buterin Proposing?
Buterin believes preparations should cover several possible AI development scenarios rather than relying on a single prediction. He proposed building infrastructure that would remain valuable regardless of how events unfold.
Among the key areas he highlighted were open technologies, cryptography, secure hardware, tools to protect against biological threats, mechanisms for verifying information, and solutions that prevent control over critical technologies from becoming concentrated in the hands of a few.
“This makes sense regardless of when superintelligence arrives,” Buterin noted.
When Should a Pause Be Discussed?
Buterin also supported the idea of defining in advance the criteria under which the development of the most powerful AI systems should be slowed down. In his view, these rules should be agreed upon before real risks emerge. The goal, he explained, is to ensure decisions are made based on pre-agreed criteria rather than during debates over how close the arrival of superintelligence may be.
The Debate Over the Future of the Labor Market
Another part of the discussion focused on how superintelligence could reshape the labor market.
One participant in the discussion under Buterin’s post argued that even the arrival of superintelligence would not lead to widespread job losses. In that person’s view, previous technological revolutions also transformed the labor market without displacing humans from the economy.
Buterin rejected that argument. He pointed out that, by definition, a superintelligence would be capable of performing any task better than a human. In that case, employers would no longer have an economic incentive to hire people for most professions.
“The opposing view only makes sense if you believe the emergence of superintelligence in the foreseeable future is unlikely,” he emphasized.