Scientist proposes using artificial intelligence to prove that humanity lives in a simulation

Do we live in a simulation? If you don’t believe it, don’t take this work into account, but if you do, or at least have doubts, follow the instructions proposed by this scientist to try to get out of it. And in them has a lot to do with the Artificial Intelligence (AI).

The Simulation Theory indicates that humanity lives inside a computer program, following the line of the saga Matrix, starring Keanu Reeves and directed by the Wachowski sisters.

The origins of this theory date back to the late 1980s, when physicist John Archibald Wheeler suggested that the universe is fundamentally mathematical, considering that it emerges from information.

Later, the philosopher Nick Bostrom, In 2003, he proposed theory as we know it today.

But what if it were true? Theoretical physicist Roman Yampolskiy indicates what the escape route would be from this hyper-realistic Matrix-style simulation.

Yampolskiy’s proposal on artificial intelligence and Simulation Theory:

“We do not evaluate the evidence for or against such a claim,” Yampolskiy points out in his paper, published last October in ResearchGate, “But we ask a computer question, namely: Can we hack the simulation?”

“More formally, the question could be formulated as: Can intelligent agents in general located in virtual environments find a way to escape from them?”

Yampolskiy, a computer scientist at the University of Louisville, makes the following proposition: force the simulators to use more and more computing power, and thus generate more power, until they can no longer ignore it.

The theoretical and computer physicist explains: “Given that the cutting-edge literature on the containment of Artificial intelligence If you answer in the affirmative (AI is unstoppable in the long run), we conclude that it should be possible to escape the simulation, at least with the help of super-intelligent AI.”

By contrast, if it is not possible to escape from the simulation, AI containment should be an important theoretical outcome for AI security research.”

Leave a Comment