(Algorithmic) Freedom

[sybokMember]

Hi,

I stumbled across a post/article
http://caitlinjohnstone.substack.com/p/censorship-by-algorithm-does-far
leading to the following pages:
http://www.jonathan-cook.net/blog/
http://twitter.com/LeeCamp/status/957346222372786176

A disturbing reading about the algorithms selecting popular/mainstream and suppressing/delaying some posts/tweets.
I do not claim to know where the truth is.
Still, there are some news that make me… worry.
I was a bit worried also when I believe that it was France announced that one of the topics of their European Union (EU) topics is to deal with so-called “hate speech”.
I recall a Star Trek TOS episode “The Savage Curtain” where Uhura replies to president Lincoln (don’t ask about the details, it was the 3rd and last season)

Abraham Lincoln: [interrupting] “What a charming negress. Oh, forgive me, my dear. I know that in my time some use that term as a description of property.”
Uhura: “But why should I object to that term, sir? You see, in our century, we’ve learned not to fear words.” [copy-pasted from IMDB.com]

How to decide where the line is between hateful and heated?
How to decide where the line is between hateful and improper choice of words to which someone is sensitive while the speaker feels/knows-of no negative connotation to them?

I would assume that the algorithms in question are to be understood as algorithms from Machine Learning (ML).
This does not mean that I think ML is bad, the opposite is true; I’ve heard about many useful applications.
Nevertheless, ML is sometimes applied in a field/way that is questionable or the model was badly trained (i.e. training data-set did not account for some of the corner cases).

Then you read stories such as that pictures of overweight (mass-challenged?) people were marked as porn because containing too much naked skin.
I’ve tried to search for a link to the news but the results of ‘obese photos incorrectly classified as porn’ were mostly NSFW and I gave up.

[sybokMember]

Free from captcha? Who would not want that.

Wait, it [DEL: may :DEL] will come with a price because it needs verify your SW is secure (enough) for you to be admitted to the privileged trusted users…
http://gabrielsieben.tech/2022/07/29/remote-assertion-is-coming-back-how-much-freedom-will-it-take/

[olsztynMember]

sybok wrote:

How to decide where the line is between hateful and heated?
How to decide where the line is between hateful and improper choice of words to which someone is sensitive while the speaker feels/knows-of no negative connotation to them?

Thank you!
Very interesting… However not surprising or unexpected. As there is no such thing as ‘public opinion’ but only ‘published opinion’, media censorship along with their ‘algorithm’ of suppressing unwanted views is often in hands of those in power. Such as recent Twitter documents reveal that governments tell the media what to highlight and what to suppress, to squash any dissent from the official propaganda of those in control.
Apart from a few pockets of resistance the so called ‘Freedom of Press’ is mostly history as is any notion of old democracy, as it has been replaced by a new form which is totalitarian in nature.

• This reply was modified 3 years, 9 months ago by olsztyn.

Live antiX Boot Options (Previously posted by Xecure):
http://antixlinuxfan.miraheze.org/wiki/Table_of_antiX_Boot_Parameters

[rokytnjiForum Admin]

Wait till AI runs things. Then you will have a reason to be upset.
http://federalnewsnetwork.com/artificial-intelligence/2022/12/white-house-looks-to-accelerate-ai-in-government-through-its-bill-of-rights/

AI For Smarter Legislation
Integrating AI into the new age of policy
Explore the ways artificial intelligence can improve both the accuracy of existing and future policies and the quality of debate about those policies in legislatures across federal, state, and local agencies.

So you EU folks don’t feel left out
http://www.vice.com/en/article/jgpb3p/this-danish-political-party-is-led-by-an-ai

Sometimes I drive a crooked road to get my mind straight.
I don't suffer from insanity. I enjoy every minute of it.
Motorcycle racing is rocket science.

Linux Registered User # 475019
How to Search for AntiX solutions to your problems

[olsztynMember]

rokytnji wrote:

Wait till AI runs things. Then you will have a reason to be upset.

Future not too far away when giant robots with AI in mind will be walking the streets and exercising deadly force on humans not authorized to walk there. San Francisco plan to deploy such robots with deadly weapons as part of police force has just been postponed after protests a few days ago and is being modified to be furnished with just explosives.
Just the beginning…

• This reply was modified 3 years, 9 months ago by olsztyn.

Live antiX Boot Options (Previously posted by Xecure):
http://antixlinuxfan.miraheze.org/wiki/Table_of_antiX_Boot_Parameters

[rokytnjiForum Admin]

Well. if the USA AI bill of rights gets passed

http://www.whitehouse.gov/ostp/ai-bill-of-rights/algorithmic-discrimination-protections-2/

protection is promised. So algorithms like used in mortgage lending , would be illegal.

Edit: Finding it weird the 2 are connected. Politics?

• This reply was modified 3 years, 9 months ago by rokytnji.

Sometimes I drive a crooked road to get my mind straight.
I don't suffer from insanity. I enjoy every minute of it.
Motorcycle racing is rocket science.

Linux Registered User # 475019
How to Search for AntiX solutions to your problems

[Brian MasinickModerator]

Wow, I haven’t been reading enough of the news! This CLEARLY sounds like the Science Fiction movies; The Terminator comes specifically to mind. The Terminators were humanoid robots capable of doing nearly everything a human does. IF you could actually get a hold of a terminator, you could reprogram them; the problem was they got out of hand and started going after humans. This is a definite risk, especially if the AI technology gets in the wrong hands.

--
Brian Masinick
Alternate "Search B":
This search page

[rokytnjiForum Admin]

Soft self repair robots make the terminator seem pretty durn close.

http://www.analyticssteps.com/blogs/self-healing-robots-new-generation-robotics

Sometimes I drive a crooked road to get my mind straight.
I don't suffer from insanity. I enjoy every minute of it.
Motorcycle racing is rocket science.

Linux Registered User # 475019
How to Search for AntiX solutions to your problems

[Brian MasinickModerator]

As for the government documents, yes, they sound like they have protections in place; so did the stuff in the Science Fiction movies. But an error caused all kinds of problems and many movies were based on it; one of the final movies in one series had one of the original inventors assist the small team that was fighting the terminators. I think they finally “defeated” the robots but not before many lives were lost and most of the area around where the story unfolded looked like a nuclear war zone.

--
Brian Masinick
Alternate "Search B":
This search page

[Brian MasinickModerator]

Again, what I’m writing about here is fiction, but the story has a very serious component to it. BEFORE any powerful AI systems are unleashed, there must be multiple levels and methods of disarming whatever systems are put in place. Otherwise, one of those fictional stories could actually come true some day.

--
Brian Masinick
Alternate "Search B":
This search page

[Brian MasinickModerator]

rokytnji wrote:

Soft self repair robots make the terminator seem pretty durn close.

http://www.analyticssteps.com/blogs/self-healing-robots-new-generation-robotics

You can say that again; that’s why I wrote my “thesis” above! 😉

--
Brian Masinick
Alternate "Search B":
This search page

[Brian MasinickModerator]

I wish we could apply our own AI robots to ward off those spammers; we seem to get the 503 messages every day around this time!
(another 5-10 minute outage during this exchange).

--
Brian Masinick
Alternate "Search B":
This search page

[olsztynMember]

Brian Masinick wrote:

the problem was they got out of hand and started going after humans. This is a definite risk, especially if the AI technology gets in the wrong hands.

Indeed…
Particularly that AI technology is controlled and implemented by those in power. Currently those in power, and actual power behind them, are not representing common people. Never have been anyway. They just provide the common people only what they need to continue as work force… If common people (workforce) step out of their way they will dealt with. AI and robots can help this task as no humans in enforcement of this task is in sight to complain…
Such vision might still look like Sci-Fi but gradually expected. Power to decide is not in the hands of common people though, to decide their future…

P.S.:
On this side of the pond it still may not look as bad as on the other side, as on the other side the common people have already been deprived of all the weapons to defend themselves against the oppressive governments…

• This reply was modified 3 years, 9 months ago by olsztyn.
• This reply was modified 3 years, 9 months ago by olsztyn.
• This reply was modified 3 years, 9 months ago by olsztyn.

Live antiX Boot Options (Previously posted by Xecure):
http://antixlinuxfan.miraheze.org/wiki/Table_of_antiX_Boot_Parameters

[mrpieceofworkMember]

Hey all, I just perused the newest comments of this thread, and I now have some thoughts, having just been exposed to the same topic within the past few hours. Bare with me… y’all appreciate this.

I recently listen to a guy who made a really good case for how AI will “do us in”, by way of something really “stupid” (like the ‘paper clip problem’) if we ever allow it to have such power. It will find the most efficient way to solve a problem, a problem that would benefit us should it be solved, and yet will inadvertently end up not “caring” for us humans. One of his examples was the AI eventually painting everything black to maximize the energy from the Sun. IOW, it won’t “do the thing” bc it hates us or whatever, it would not consider us at all. Something to consider, indeed.

Then, a little while later, I came across a website that has people submit images that have the entire gamut of all 16 million or so RGB color combinations, each color point only being one pixel, and never repeating any. IDK understand how they do it, I assume through custom coded parsing scripts/algorithms. Some of the images are of famous paintings and such, really neat stuff. (allrgb.com)

Someone asked that ChatGPT AI to come up with a solution…

and it returned a “grey” image (all the colors evenly distributed)

THAT’S our AI future… a grey blob. LOL

[sybokMember]

One “minor” issue of AI controlled life including quality of debates etc. is that society will homogenize and get poorer as far as thinking is concerned.

Because everything that deviates will be suspicious (like solving captcha too fast) and people will learn to express/behave in a way that is not penalized (I slowed down my solving them and now I even do not think about it).
Less troubles, less diversity… less richness in thinking, diminished (average) ability to solve problems (because all/most people might tend to thing the same way thus diminishing competing/alternate points of view).

This is a similar conclusion as @mrpieceofwork pointed out – a grey blob… of unified (easy to [DEL: control :DEL] govern) society.
Something that is already in progress in studio art for the masses (striving too hard not to offend anyone), see e.g.
http://www.youtube.com/watch?v=F2ngB-zjVmM
http://www.youtube.com/watch?v=jXCAAZ8_fCM

Hopefully, it will not come that far.
Our mission, should we choose to accept it, is to be (reasonably) alert and support critical points of view to raise awareness of the downsides and try to block similar tendencies (by not buying/using products, by signing petitions etc.).

PS: And there’s room for bugs… which reminds me of a cartoon-joke about a code error; I was unable to find it [the joke, not the bug] but the question (and related discussion) in the link summarizes (and explains) the joke fair enough
http://www.reddit.com/r/xkcd/comments/52xh1y/can_someone_help_me_find_a_specific_xkcd/

[Author]

Posts