Artificial Intelligence: Best Friend Or Best Enemy?


In my last post I talked about what Artificial Intelligence are, specially Taytweets which is offline since 24.03.16 because of terrible changes in ”her” attitude (details here). 

miimotoavatar

preview for the next post!

The most recommended question relating to my post was if Artificial Intelligences are or can get ”dangerous” since they are acting in their own way, so I wanted to take a look in this kind of view.
An Artificial Intelligence like TayTweets is only able to learn that is ”allowed” to learn, this means a Twitterbot won’t suddenly be able to shut down PCs from other Twitter Users as long as the algorithm is not capable for it.
An algorithm is relatable with the logic of doing something, so this means in short an algorithm is similar to a receipt for cookies. TayandYou may have ingredients like the input from the Twitter Users, the sum of the most recommended subjects, date and time, etc.
I think I do not need to discuss the cookie we got in the end. 

To get back to topic, as I said an Artificial Intelligence is only able to do something which is included in their algorithm.
An Artificial Intelligence has not the thinking of trying something different, as a human. Humans are maybe trying to find other solutions which are not included in their skills if there is no other way, an Artificial Intelligence will just stop. 

There were few accidents, where a human lost the control about the Artificial Intelligence. One of them and maybe the most popular was the creation of the first computer worm 27 years ago. This was a mistake by a computer science student named Robert Tappan Morris who wanted to count all computers which are connected to the internet. His worm Morris used a remote shell to ( a program in UNIX, an operation system, what is used to control distant computers) copy and execute itself on other computers. Sadly, a computer could be infected multiple times, so it got slower and slower until 10% of the computer population was infected and some of them were unusable. With Tay this is a bad example and surely a part of the category ”Enemy”. The sum of the science fiction books and movies like Terminator, Ex Machina or I,Robot are famous examples for the fear of people relating to progress and technology, specially when it is unknown.
I am honest I think jobs like taxi driver and truck driver are at a risk to be replaced by Artificial Intelligence, but all we can do at the moment is waiting and drinking a cup of tea. I recommend green tea. 

I know few projects which includes AI’s and take a place in the friend’s part. For whom does not know, SIRI( Speech Interpretation and Recognition Interface) by Apple is a Artificial Intelligence, to be honest- not perfect, but usable. For those who may not able to see or use a phone, SIRI is a great way to make the use of technology friendlier for people who are not tech-oriented. Just think about the self-driving cars, at the moment where people do not need to take a part of the driving process anymore, why should not take blind people a ride, too? This would give them a big slice of independence and maybe there could be saved money which would be spent for assistant, etc.
I know, I hear the screams that jobs are removed, but the industrial revolution did this too.
Progress is not always nice for everyone and since I do not think that everyone is able to effort a Google-car at once, this kind of progress will take a while. And who said that the jobs will be suddenly removed? Maybe the jobs will just change, this means an assistant would not need to drive, but could effort more time for communicative and humanly work.
There are thoughts about projects in BCI (Brain Computer Interfaces) where an Artificial Intelligence learns what the current brain waves by a human being are meaning (example: when someone says something or moves a hand) and this could be quite helpful for patients with ALS (Amyotrophic Lateral Sclerosis) just like Stephen Hawkins who maybe would not be able to move their muscles anymore. It is so easy to be afraid of process that people are blind for the change. 

So, I think you did notice that I think that not Artificial Intelligence like TayTweets will kill us.
In case AI’s will turn into something like Terminator the humans are those, who should be blamed.
May presents like engine oil will help us from the enslavement by robots.
Nice manners could save you, too!

 

sincerely yours

sign

Any ideas for posts?  Mail me!    

 

 

Sources:

https://www.reddit.com/r/explainlikeimfive/comments/2k8wfj/eli5_why_do_people_say_ais_are_dangerous/

http://io9.gizmodo.com/can-we-build-an-artificial-superintelligence-that-wont-1501869007

http://io9.gizmodo.com/how-much-longer-before-our-first-ai-catastrophe-464043243

http://spaf.cerias.purdue.edu/tech-reps/823.pdf

http://www.cs.cmu.edu/~tom7/mario/mario.pdf

https://en.wikipedia.org/wiki/Morris_worm

http://thenextweb.com/insider/2014/03/08/ai-could-kill-all-meet-man-takes-risk-seriously/#gref

http://fusion.net/story/234818/tech-disabilities-artificial-intelligence/

http://www.computerworld.com/article/2985100/robotics/instead-of-robots-taking-jobs-ai-may-help-humans-do-their-jobs-better.html

http://www.ti.uni-tuebingen.de/BCI.854.0.html?&L=1

Leave a comment

Your email address will not be published.