Theoretically, it is impossible for humans to control a super-intelligent Artificial Intelligence (AI) machine. It is also nearly impossible to detect when such an AI is created.

In 2015, Google developed AI software that was able to play – and regularly win – classic arcade games. Their findings were detailed in the journal Nature. In 2018, a study published in Science outlined how AI can regularly master chess, shogi (Japanese version of chess), and Go (ancient Chinese strategy game) through self-play. Smartphone apps can easily beat world chess champions such as Magnus Carlsen.

The Three Laws of Robotics are:

  1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.
  2. A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
  3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws.

Additionally, the “Zeroth Law” states that a robot may not harm humanity, or by inaction, allow humanity to come to harm.

Researchers from Tufts University’s Human-Robot Interaction Lab are trying to develop mechanisms for robots to reject orders received from humans, as long as they have a legitimate reason to do so.

Now, researchers are warning that any program written to stop AI from harming humans (and destroying the world while they’re at it) might conclude to halt or not. Mathematically, it is impossible for us to be absolutely sure which route the AI will take.

An alternative would be to teach the AI “ethics” of sorts and order it to not destroy the world. Researchers say this limits the super-intelligence capabilities and would force the AI to be cut off from parts of the internet or specific networks. Moreover, this defeats the purpose of creating AI in the first place: If the AI isn’t going to be used to solve problems beyond a human scope, then why exist in the first place?

As AI research expands, we might not recognize when the super-intelligence beyond human control forms. It would be incomprehensible, thus, uncontrollable.  

Read the full article in the Journal of Artificial Intelligence Research to learn more. Check out a full paper that outlines how humans are developing mechanisms to allow AI to reject any given orders.

Founded in 1994, Creative Programs and Systems provides professional results for all computer needs. We design, create, and code an array of custom software programs and websites; implement internet marketing strategies for enhanced Search Engine Optimization (SEO) results; repair and provide support for computers of both residential and professional nature; build custom systems and servers, and offer secure data backups. Need assistance or want to learn more? Call us at 810-224-5252 or e-mail info@cpsmi.com.

Written by the digital marketing staff at Creative Programs & Systems: www.cpsmi.com.

Leave a Reply