What can't AI do - Humans Can

May 25, 2021  7572 seen

What can't AI do - Humans Can

Artificial Intelligence

Without a doubt, artificial intelligence has made tremendous changes in several areas of human life. In fact, artificial intelligence is no longer an interesting topic for thinking about the future; it's here! A recent future of employment study predicts that nearly 50% of the jobs in the U.S. are likely to be automated by 2030. Here are some things an AI can’t do now.

 

Solve Annoying Interview Puzzles

Perhaps you don't want artificial intelligence to solve any problem. Maybe you just want AI to do some useful things such as play computer games or calculate your expenses. A Russian mathematician Yuri Matiyasevich used Turing’s theory to show that computers can’t even solve simple interview puzzles.
Anthony has some marbles and gives half to Sam and a third to Angela, how many marbles did Anthony have if…

Solving these types of problems with an algorithm known as Hilbert’s 10th problem nowadays is impossible.

Solve Annoying Interview Puzzles

 

Write Bug-Free Software

Another challenging area for artificial intelligence is finding bugs in software. Don't get it wrong, artificial intelligence can be of great help in finding patterns that might indicate a bug, but Gordon Rice's little-known theory expands on Turing's original proof to show that computer programs cannot know anything interesting about another computer program.. computer program. Thus, the AI ​​cannot know if a computer program is free from errors, if it will have a blue screen, or if it will have an out-of-memory error. Indeed, one computer program cannot tell you anything non-trivial about another, and since AI is nothing more than computer programs, it cannot help you write bug-free software.

 

Make Moral Choices

Unlike humans, who sometimes can make instantaneous moral choices and then analyze them after the fact to fit the rules, artificial intelligence only uses rules. According to Kurt Gödels’ theory, any time you have a system with formal rules there will be problems those rules can’t answer –  the system will be incomplete. So, while computers obey the law - and we'd better hurry up and program them all with Isaac Asimov's three laws of robotics - they can't handle new situations.

 

Exercise Free Will

Artificial intelligence based on digital computers must follow the rules of its program. These rules are deterministic and do not allow any free will. Based on Conway and Cochen theory, fundamental particles in our universe do indeed have free will, and humans tap into this basic free will so that their actions are not predetermined. Can we use this fact to give our machines free will? We will, but when we do, they will no longer be AI, they will be Real I.

 

Make a Person Feel Heard

Working in customer service is like being a psychologist. When a customer contacts you with a problem, the problem is not always your product or service. Sometimes all customers need is just some attention. It's not always about selling something. For example, the longest call in the call center with a customer was about 10 hours.