- Jul 27, 2015
- 5,458
Quote : " Automation and artificial intelligence are already replacing human judgement. By using smart devices, we give more and more power to artificial intelligence -- and it will soon make dangerous choices. "
Dangerous Choices
- Thread starter upnorth
- Start date
Quote : " Automation and artificial intelligence are already replacing human judgement. By using smart devices, we give more and more power to artificial intelligence -- and it will soon make dangerous choices. "
- Jul 6, 2017
- 271
Great video with some interesting questions raised. Together with implementation of great technological developments, unfortunately often come some not so great aspects which pose serious risks of accidents, misuse and abuse. For example, progress made in the field of genetic engineering, which is extremely useful in medicine, can also be misused for biological warfare. Likewise, IoT has ability to make our everyday life much easier and convenient, but at the same time it opens vulnerabilities which could be used to turn our lives into a hell.
Thanks for sharing, I really enjoyed watching it.
Thanks for sharing, I really enjoyed watching it.
- Jul 27, 2015
- 5,458
Thanks for mention this as I fully agree but it's a also a very real and potential threat when it comes to food and beverages. The medicine part is a particularly nasty and very scary one as to reports I seen several years ago it was already out of control in some parts of the world. Those kind of " malware " kills people but it's another topic.
The Beam I found very interesting and also funny and guess it has potential as I seen kids in my country using similar devices.
Last edited:
- Jul 6, 2017
- 271
Genetic engineering is only one example, the same observation can also be applied to any other technology. It's a hot topic and there are lot of papers with interesting points of view, especially on the subject of AI (notion of intelligence explosion which leads to technological singularity might be the most entertaining one ). Anyhow, what I wanted to say is that understanding that not every technological development necessarily leads to a social progress and recognizing the potential dangers are the crucial steps in preventing disasters and minimizing different kinds of risks.
). Anyhow, what I wanted to say is that understanding that not every technological development necessarily leads to a social progress and recognizing the potential dangers are the crucial steps in preventing disasters and minimizing different kinds of risks.
- Jul 5, 2016
- 416
This new technology will be added to AI I am guessing. Scientists have designed virtual human brain cells
Skynet is coming faster than you think...
Thanks for sharing, really interesting
The current efforts in areas such as deep computational learning involve algorithms that use large amounts of information. The software is not necessarily programmed to “know” the rules before, but to find the rules or it is prepared to be guided by the rules.
It is something amazing, but it is not clear if this is a path that has equivalence with the way how humans think.
I remember Stephen Hawking saying that would be the ability of a strong artificial intelligence to “evolve” much more quickly than biological systems by draining our resources.
I think that it is a good hypothesis and maybe it will be a threat. The threat of artificial intelligence is not that it will be sadistic or devastating (unless we deliberately or recklessly, do it in a way that it is so), but it will follow its own evolutionary system.
The current efforts in areas such as deep computational learning involve algorithms that use large amounts of information. The software is not necessarily programmed to “know” the rules before, but to find the rules or it is prepared to be guided by the rules.
It is something amazing, but it is not clear if this is a path that has equivalence with the way how humans think.
I remember Stephen Hawking saying that would be the ability of a strong artificial intelligence to “evolve” much more quickly than biological systems by draining our resources.
I think that it is a good hypothesis and maybe it will be a threat. The threat of artificial intelligence is not that it will be sadistic or devastating (unless we deliberately or recklessly, do it in a way that it is so), but it will follow its own evolutionary system.
- Jul 6, 2017
- 271
What you @tim one said reminded me of an article in which Hawking states that creating AI might be the biggest mankind accomplishment (but it could also be it's last). He also argues that everything that our civilization has achieved is a product of human intelligence (and I'd say not only intelligence but also our emotions, our sex drive, and damn it - even our madness at times, or any other aspect that makes human race what it really is) so trying to predict what could be achieved if AI is added into that equation is an extremely hard challenge.
Agree, and I think an engineering challenge could be the need of a system with the complexity of the human brain to support the AI, but of course the brain has co-evolved in a natural way with our intelligence. So it's a bit stupid to think to sit down and design the ideal circumstances for the birth of a new type of intelligence, given that we don't know exactly what should be these circumstances.
Or maybe we are really creating those conditions. The machine learning can be only a piece of the puzzle of artificial intelligence, but what can happen when this intelligence lives through extensions of the Internet? Algorithms for the transport of data packets, sensors, medicine, genetic engineering, and a myriad of “pieces” that speak to each other in a direct or indirect way...
- Jul 6, 2017
- 271
That's good point and I agree with what you said.
There are still a lot of question marks left and technological aspect is just one of many pieces of puzzle in that picture. If you're interested in subject you might want to check these two papers out:
Armstrong, Bostrom & Shulman (2013) Racing to the Precipice: a Model of Artificial Intelligence Development
http://www.fhi.ox.ac.uk/wp-content/...el-of-artificial-intelligence-development.pdf
Bostrom (2006) Technological Revolutions: Ethics and Policy in the Dark
https://nickbostrom.com/revolutions.pdf
Edit:
Shulman & Bostrom (2012) How Hard is Artificial Intelligence? Evolutionary Arguments and Selection Effects
https://nickbostrom.com/aievolution.pdf
I actually meant to put this one and not the one on Technological Revolutions, but I'll leave them all here in case anyone wants to check them allaout
There are still a lot of question marks left and technological aspect is just one of many pieces of puzzle in that picture. If you're interested in subject you might want to check these two papers out:
Armstrong, Bostrom & Shulman (2013) Racing to the Precipice: a Model of Artificial Intelligence Development
http://www.fhi.ox.ac.uk/wp-content/...el-of-artificial-intelligence-development.pdf
Bostrom (2006) Technological Revolutions: Ethics and Policy in the Dark
https://nickbostrom.com/revolutions.pdf
Edit:
Shulman & Bostrom (2012) How Hard is Artificial Intelligence? Evolutionary Arguments and Selection Effects
https://nickbostrom.com/aievolution.pdf
I actually meant to put this one and not the one on Technological Revolutions, but I'll leave them all here in case anyone wants to check them allaout
Last edited:
Thanks so much for sharing those papers, I am really interested in that!
- Jul 6, 2017
- 271
Thanks so much for sharing those papers, I am really interested in that!
You're welcome.
I edited the previous post with another link. There are some interesting papers on that subject written by other authors like, for example, Good's Speculations Concerning the First Ultraintelligent Machine written in 1960's (!) or some papers by Kurzweil, Vinge or Chalmers. I think they might be available for free, but if you'll have problems finding them, let me know and I can check if I still got some of those papers.
These good reads will be with me this night for sureYou're welcome.
There are some interesting papers on that subject written by other authors like, for example, Good's Speculations Concerning the First Ultraintelligent Machine written in 1960's (!) or some papers by Kurzweil, Vinge or Chalmers. I think they might be available for free, but if you'll have problems finding them, let me know and I can check if I still got some of those papers.
This topic is part of my study path, or better I will deal with it the next year I think, but in the meantime I'm reading a lot, especially online posts but your papers are really exstensive!
- Jul 6, 2017
- 271
These good reads will be with me this night for sure
I'm glad to hear that and good luck with your studies.
This subject was somewhat connected with what I chose to research on my final year so later today I'll go through my external disks and check if I still have some literature that you might find useful. If I do find something, I'll let you know. Thank you againI'm glad to hear that and good luck with your studies.
- Apr 28, 2015
- 9,102
@Prorootect: try to post with a higher font size
The way I see it... Robots are made by humans and thus robots are prone to mistakes. Artificial Intelligence systems depend on humans for the development and/or maintenance, also making them prone to mistakes too!
A robot may make another robot, or an Ai system may be clever enough to handle development plans for a new Ai... But the original was still dependent on the human.
A robot may make another robot, or an Ai system may be clever enough to handle development plans for a new Ai... But the original was still dependent on the human.
- Apr 28, 2015
- 9,102
@Prorootect: Your last post, in default settings and font size in my FF57, looks like this:
Your font size is lower than standard/normal font size of the rest of posts in general, difficult to read
Your font size is lower than standard/normal font size of the rest of posts in general, difficult to read
- Aug 31, 2014
- 655
Last edited:
Similar threads
