Dr. Roman Yampolskiy, a leading AI safety researcher, has issued a stark warning: humanity is not ready for the rapid advancement of artificial intelligence, with superintelligent systems potentially arriving within two years. In a recent interview, Yampolskiy outlined a timeline where artificial general intelligence (AGI) transitions to superintelligence, bringing unprecedented risks.
Yampolskiy argues that current efforts to embed ethics into AI are fundamentally flawed, because we cannot fully specify or control what a superintelligent system will do. He points to the broken career ladder as automation already displaces jobs, and warns that free AI labor could concentrate wealth further.
The researcher predicts that within five years, AI systems will surpass human capabilities in virtually all domains, making it impossible to reverse or contain them. He calls for a global pause on advanced AI development, but admits that competitive pressures make this unlikely.
Yampolskiy's warnings echo those of other experts who advocate for stringent safety measures before deploying increasingly powerful AI models.