SIGforum.com    Main Page  Hop To Forum Categories  The Lounge    Black Mirror AI - we won't be able to trust this stuff
Page 1 2 3 4 5 6 7 
Go
New
Find
Notify
Tools
Reply
  
Black Mirror AI - we won't be able to trust this stuff Login/Join 
Peace through
superior firepower
Picture of parabellum
posted
And one day, we'll have to fight the robots. It's inevitable.

This clip is worth your time, I think. It gets very interesting towards the end.

 
Posts: 108373 | Registered: January 20, 2000Reply With QuoteReport This Post
Step by step walk the thousand mile road
Picture of Sig2340
posted Hide Post
I already don’t trust AI.

It is technology that is ripe for misuse, and as we know, if it can be misused it will be misused. And not just Rule 34, misuse either.

Plus the whole SkyNet deal.





Nice is overrated

"It's every freedom-loving individual's duty to lie to the government."
Airsoftguy, June 29, 2018
 
Posts: 31650 | Location: Loudoun County, Virginia | Registered: May 17, 2006Reply With QuoteReport This Post
Member
Picture of Rotty37
posted Hide Post
Possibly replacing Google altogether. I find it interesting that Chat GPT is by Open AI, in which Elon Musk is involved in some form. Interesting.


ΜΟΛΩΝ ΛΑΒΕ
 
Posts: 4452 | Registered: December 27, 2005Reply With QuoteReport This Post
Back, and
to the left
Picture of 83v45magna
posted Hide Post
AI is a scary proposition and there is no way in hell humankind can avoid it's development and proliferation.

This is an AI that is supposed to render unbiased, logical, factual answers to questions. Is it actually surprising that leftists would seize any opportunity to bake their opinions and preferences into it?
 
Posts: 7316 | Location: Dallas | Registered: August 04, 2011Reply With QuoteReport This Post
Wait, what?
Picture of gearhounds
posted Hide Post
^^^
This will be the true problem with AI- the manipulation by powers that seek to control the population, not that it will take on a life of its own and try to destroy humanity.

I’d be interested to hear its response to something like “why is the second amendment important to Americans?”. I’m thinking we wouldn’t be comfortable with the answer.




“Remember to get vaccinated or a vaccinated person might get sick from a virus they got vaccinated against because you’re not vaccinated.” - author unknown
 
Posts: 15722 | Location: Martinsburg WV | Registered: April 02, 2011Reply With QuoteReport This Post
Run Silent
Run Deep

Picture of Patriot
posted Hide Post
It's no different than liberal parents raising a liberal child with liberal views.

Whomever programs the AI is the "parent" and can therefore program in the parents bias.


_____________________________
Pledge allegiance or pack your bag!
The problem with Socialism is that eventually you run out of other people's money. - Margaret Thatcher
Spread my work ethic, not my wealth
 
Posts: 7002 | Location: South East, Pa | Registered: July 04, 2002Reply With QuoteReport This Post
Member
Picture of myrottiety
posted Hide Post
AI & Machine Learning is some of the fastest growing space in technology.

It's just in it's infancy so far. Computer Vision is another place that's come a long way & is a bit scary as far as applications.




Train how you intend to Fight

Remember - Training is not sparring. Sparring is not fighting. Fighting is not combat.
 
Posts: 8889 | Location: Woodstock, GA | Registered: August 04, 2005Reply With QuoteReport This Post
Optimistic Cynic
Picture of architect
posted Hide Post
How many AI's do we have here on SF? Some of you guys I'm not so sure about....

Joking aside, I agree that this is a very significant development. I recently had a CT scan performed where the results came back so fast (minutes) that I am convinced that they had to have been generated by an AI rather than a human doctor. This didn't seem to matter to my gastroenterologist.

When AIs can reliably and unfailingly pass a Turing test, it raises a lot of questions. Can they be considered intelligent beings? What rights do they have? As a defender of human supremacy, how do I determine that I am talking to another human rather than an AI? If AIs are prohibited from participating in certain activities, how would you tell?
 
Posts: 6622 | Location: NoVA | Registered: July 22, 2009Reply With QuoteReport This Post
Still finding my way
Picture of Ryanp225
posted Hide Post
I already don't trust ai algorithms that are used for target marketing and most likely other stuff.
It's only going to get worse at a large order of magnitude going forward.
 
Posts: 10851 | Registered: January 04, 2009Reply With QuoteReport This Post
His Royal Hiney
Picture of Rey HRH
posted Hide Post
quote:
Originally posted by Sig2340:
I already don’t trust AI.

It is technology that is ripe for misuse, and as we know, if it can be misused it will be misused. And not just Rule 34, misuse either.

Plus the whole SkyNet deal.


What Elon Musk is showing what happened at Twitter which mirrored the other social media tech companies show that who controls the code controls how you think.



"It did not really matter what we expected from life, but rather what life expected from us. We needed to stop asking about the meaning of life, and instead to think of ourselves as those who were being questioned by life – daily and hourly. Our answer must consist not in talk and meditation, but in right action and in right conduct. Life ultimately means taking the responsibility to find the right answer to its problems and to fulfill the tasks which it constantly sets for each individual." Viktor Frankl, Man's Search for Meaning, 1946.
 
Posts: 19841 | Location: The Free State of Arizona - Ditat Deus | Registered: March 24, 2011Reply With QuoteReport This Post
thin skin can't win
Picture of Georgeair
posted Hide Post
quote:
Originally posted by Patriot:
It's no different than liberal parents raising a liberal child with liberal views.

Whomever programs the AI is the "parent" and can therefore program in the parents bias.


Not entirely accurate for "artificial general intelligence", which is a fancy way of saying roughly equivalent to being conscious. "Programming" is just the on ramp for startup, but is unlikely to be binding.

Once the lights come on, the control problem is infinitely harder to envision or control by how the lead up development takes place than it seems. Then there's the issue of multiple competing AGI projects that will likely be underway at the same time coupled with lack of awareness or agreement on controls or seeds. All that before you get to the issue of takeoff trajectory of AGI and if it slowly evolves (unlikely) or rockets up a pathway of exponential growth and pursuit of whatever objectives are set.

If you want to read a good book on this, I recommend Superintelligence: Paths, Dangers, Strategies as a starter, but even that at 8 years old is a bit out of date. Others out there with similar but updated views.

One of the examples he uses is if you develop a true AGI with the simple objective of "make as many paper clips as possible" in effort to avoid Skynet scenario, then all else is irrelevant. That could lead to the consumption of all resources on Earth and any other planets/moons/etc. the AGI can reach forever all in the pursuit of as many paper clips as possible.

Personally, I've got enough paper clips to last me a lifetime already, so pick better folks!

Most of what folks are calling AI now is really just machine learning rather than AGI. Very different.

This message has been edited. Last edited by: Georgeair,



You only have integrity once. - imprezaguy02

 
Posts: 12527 | Location: Madison, MS | Registered: December 10, 2007Reply With QuoteReport This Post
delicately calloused
Picture of darthfuster
posted Hide Post
We are borg. You will be assimilated. Resistance is futile.



You’re a lying dog-faced pony soldier
 
Posts: 29821 | Location: Highland, Ut. | Registered: May 07, 2008Reply With QuoteReport This Post
Member
Picture of Ozarkwoods
posted Hide Post
Interesting piece. Meld the minds of the youth, and you will change a nation.


ΜΟΛΩΝ ΛΑΒΕ
 
Posts: 4874 | Location: SWMO | Registered: October 20, 2009Reply With QuoteReport This Post
Official forum
SIG Pro
enthusiast
Picture of stickman428
posted Hide Post
Could the right group of people build an AI bot that is truly unbiased and thus able to give one truly objective arguments from multiple sides of any argument?

It seems like the main problem with AI comes from the flaws of its creator and parent company. Would the right group of individuals be a solution?


~~~~~~~~~~~~~~~~~~~

The price of liberty and even of common humanity is eternal vigilance
 
Posts: 21185 | Location: San Dimas CA, the Old Dominion or the Tar Heel State…flip a coin  | Registered: April 16, 2007Reply With QuoteReport This Post
Peace through
superior firepower
Picture of parabellum
posted Hide Post
Fascinating

We need to pay attention to this stuff.

GIGO: garbage in, garbage out

 
Posts: 108373 | Registered: January 20, 2000Reply With QuoteReport This Post
Optimistic Cynic
Picture of architect
posted Hide Post
quote:
Originally posted by stickman428:
Could the right group of people build an AI bot that is truly unbiased and thus able to give one truly objective arguments from multiple sides of any argument?

It seems like the main problem with AI comes from the flaws of its creator and parent company. Would the right group of individuals be a solution?
History seems to have shown that this is not likely.

More likely, IMO, is that an unbiased (from a human PoV) AI will only arise multiple generations after AI's start replicating themselves. Then, of course, they will develop their own philosophy and PoV that will not be congruent with humanity's. Whether or not this will be for the net benefit of the human race is hardly certain. Would horses exist in the world today if humans didn't feel they were of more benefit as work animals than as a food? Successive clashes of moralities seems inevitable.
 
Posts: 6622 | Location: NoVA | Registered: July 22, 2009Reply With QuoteReport This Post
Ammoholic
Picture of Skins2881
posted Hide Post
Skynet is probably already self aware.






Jesse

Sic Semper Tyrannis
 
Posts: 20962 | Location: Loudoun County, Virginia | Registered: December 27, 2014Reply With QuoteReport This Post
Member
posted Hide Post
I am forced to use AI to search documents. It is not intuitive and is just plain slow. Not ready for prime time. In an earlier post I mentioned the robot floor cleaner needs improvement. A loud beep is just not the same as excuse me. Of course the STEM folks seem to love this shit. Another way to limit human interaction.
 
Posts: 17381 | Location: Stuck at home | Registered: January 02, 2015Reply With QuoteReport This Post
Member
Picture of Orthogonal
posted Hide Post
After reading the news at the Register regarding Google's chat bot business dilemma I was tempted to sign on to Google and ask if it had a proof or disproof for Riemann's Hypothesis. But I didn't so that's still open. I really wanted to break it.

If you wish to know about either see here

https://www.theregister.com/20...n_brief_ai/?td=rt-3a
and
https://empslocal.ex.ac.uk/peo.../zeta/riemannhyp.htm
Big Grin
 
Posts: 519 | Registered: May 03, 2003Reply With QuoteReport This Post
Member
Picture of maladat
posted Hide Post
quote:
Originally posted by Georgeair:
Most of what folks are calling AI now is really just machine learning rather than AGI. Very different.


As far as stuff that is actually happening and getting discussed in popular media, machine learning is pretty much it.

AI-as-a-subfield-of-Computer-Science includes a bunch of other stuff, but the other stuff isn’t dramatic and exciting to the average non-Computer-Scientist and gets very little public attention.

AGI/strong AI is still very firmly in the realm of science fiction.
 
Posts: 6319 | Location: CA | Registered: January 24, 2011Reply With QuoteReport This Post
  Powered by Social Strata Page 1 2 3 4 5 6 7  
 

SIGforum.com    Main Page  Hop To Forum Categories  The Lounge    Black Mirror AI - we won't be able to trust this stuff

© SIGforum 2024