Artificial Super Intelligence (ASI)

(The Louvre Museum, France - Ching-Fuh Lin)

Artificial Super Intelligence (ASI) is one of three types of artificial intelligence. ASI is a hypothetical AI, i.e. we cannot achieve it, but we know what would happen if we achieved it. So basically it's imaginary artificial intelligence that not only explains or understands human behavior and intelligence, but ASI is where machines will become self-aware/self-alert enough to surpass human intelligence and behavioral capabilities. With superintelligence, machines can think of possible abstractions/interpretations that humans simply cannot think of. This is because the human brain's ability to think is limited to billions of neurons.

ASI is artificial intelligence that surpasses human intelligence and capabilities. It's the best in everything - math, science, medicine, hobbies, you name it. Even the brightest human minds can't come close to the capabilities of super artificial intelligence. Among the types of artificial intelligence, super artificial intelligence is what most people refer to when they talk about robots taking over the world. Or about AI overthrowing or enslaving humans. (Or most other sci-fi AI tropes.) But rest assured, ASI is purely speculative at this point. That said, it's unlikely to be around for long (if at all).

By contrast, a robust AI system will have human-level capabilities in a wide range of domains, like most humans, and be able to find solutions to new problems when faced with them. Strong AI (AGI) has yet to be proven, and many experts doubt it might even exist. ASI will far outperform powerful AI.