AI is not coming, "THEY'RE HERE!"

[rockedge]

@jamesbond The difference is in the hardware. Better processors more connections and Quantum computers. Hardware switches that behave like neurons and synapses.

The biggest breakthroughs are not the thing called "AI" but it is the advances in abilities to translate brain signals and interact with them including generating responses. Those technologies have HUGELY advanced, and very few speak of it unless you are directly involved with the work or are a family member of the amputees and quadriplegics that are participating in the research programs.

They now can hook up a brain to a machine arm and the subject can move that prosthetic hand and arm like it was the limb he was born with. These advances also have led to the ability to successfully bypass damaged nerve connections and use a hardware solution to give motion to someone paralyzed in an accident from the neck down and most importantly can now return feeling and the sense of touch on the artificial hand and fingers that are directly processed by the human brain.

This "Bionic" advancement later matched to improved software that works on neural networks artificial and natural, will be the verge of a real potential problem once mankind is added to the mix. Once it all operates well enough it will be the base for new military weapons.

It all boils down to either it's use as weapon to kill and control or to make ridiculous sums of money as in profit. Last thing will be the medical advances that it all began with.

[sonny]

Will Future AI Systems be Legally Liable?

1. Malfunctions in accidents of self-driving cars
2. Violation of intellectual property rights for authors of texts
3. Assigning wrong prescriptions from faulty diagnosis of medical data
4. Creating pandemics by genetically engineering viruses
5. Controlling weapons of mass destruction
6. Faking scientific results that damage our quality of life

https://avi-loeb.medium.com/will-future ... c4339da547

[backi]

@sonny wrote:

Will Future AI Systems be Legally Liable?

Good Questions .
Rather complex Topic.

[pp4mnklinux]

ChatGPT answer:

"bionicpup@gmail.com

Q~Will Future AI Systems be Legally Liable?

A~The question of whether future AI systems will be legally liable is a complex and highly debated issue. As of now, AI systems are generally treated as tools or instruments, and any legal liability rests with their operators, designers, or manufacturers. However, as AI systems become more advanced and autonomous, it is possible that they will become more responsible for their actions and may be held liable for any harm they cause.

There is currently no legal framework in place for holding AI systems accountable for their actions, and this is a significant challenge that will need to be addressed as AI becomes more prevalent in society. Some experts have proposed creating a new legal category of "autonomous agents" that would be held responsible for their actions, but this is still a matter of ongoing debate.

It is important to note that even if AI systems are held liable for their actions, the question of how to assign responsibility and determine damages will be complex. This is because AI systems are often the result of complex algorithms and data inputs, making it difficult to determine exactly what caused a particular outcome.

Overall, it is still too early to say with certainty whether future AI systems will be legally liable, but it is clear that this is an important issue that will need to be addressed as AI continues to advance"

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

Will Future AI Systems be Legally Liable?

1. Malfunctions in accidents of self-driving cars
2. Violation of intellectual property rights for authors of texts
3. Assigning wrong prescriptions from faulty diagnosis of medical data
4. Creating pandemics by genetically engineering viruses
5. Controlling weapons of mass destruction
6. Faking scientific results that damage our quality of life

https://avi-loeb.medium.com/will-future ... c4339da547

[Clarity]

There's only 1 other member on this thread who has actually taken a moment to get it to see what all the fuss is about.

The naysayers may see something enlightening.

OR if you are near college students (particularly IT students) they can help you get started for your personal evaluation of all the 'fuzz' or FUD.

Enjoy

[Grey]

OR if you are near college students (particularly IT students) they can help you get started for your personal evaluation of all the 'fuzz' or FUD.
Enjoy

I am not far from one institute. I talked to the students on the bench. Here is their idol and a sample of the future of Artificial Intelligence. Smart, intelligent (when he really needs it), loves humans and is always ready to come to the rescue.
And in extreme cases (if the scholarship is small), students can take him to the reception point of ferrous and rare earth metals and earn money for their studies. Wow, they still teach them something at the Faculty of IT
I enjoyed it:

Bender.jpg (28.74 KiB) Viewed 8755 times

[geo_c]

There's only 1 other member on this thread who has actually taken a moment to get it to see what all the fuss is about.

The naysayers may see something enlightening.

OR if you are near college students (particularly IT students) they can help you get started for your personal evaluation of all the 'fuzz' or FUD.

Enjoy

True 'dat.

Just like I won't be checking out what all the fuss is about when "Musk's" brain implant technology finally gets off the ground.

[backi]

@geo_c wrote:

Just like I won't be checking out what all the fuss is about when "Musk's" brain implant technology finally gets off the ground.

https://eu.usatoday.com/story/tech/2022 ... 811520002/
Elon Musk's Neuralink wants people to control computers with their minds."(or is it the other Way around?)" How close are they?

Elon Musk has plenty of ambitious plans for the world: Trips to Mars via SpaceX. An "everything app" through Twitter. Brain implants that can transmit human thoughts via Neuralink.

Musk's brain implant company in search of human trials partner:
https://www.reuters.com/business/health ... 023-03-27/

Any Volunteers here on this Forum?

[Grey]

Any Volunteers here on this Forum?

I thought we were in the same boat and fighting with AI. And you are recruiting volunteers for experiments. Perhaps the chip is already inside you Oh, no. We have to apply protocol number 999.

[backi]

@Grey wrote:

We have to apply protocol number 999.

Sure.....We`re living in an Upside Down World

Is this (protocol number 999) some kind of Antidot or an inverting/protecting magical Ritual to neutralize the occult Power/Influence of "The Mark of the Beast-666).....????........

Cyber Satan?????

.....just asking for a Friend.....

Recently, many claim that the implant of medical or financial electronic ID chips are the "Mark of the Beast-666".

Others point to credit or debit cards. While those items may be an indication of what is to come, Bible scholars agree that the "Mark of the Beast" will be a recognizable sign of those who have voluntarily chosen to follow the Antichrist.

In the book of Revelation, there are several passages that describe the Antichrist as a beast arising from the sea, who will mark his followers, with signs upon their hands and forehead. These Bible verses include descriptions of the Antichrist’s appearance, his power, and his attempt to rule over the world.

Book of Revelation 13:17 New King James Version (NKJV)

and that no one may buy or sell except one who has the mark or the name of the Beast, or the number of his name.(which is 6-6-6)

Hey man.....Things are getting stranger and stranger..... Day by Day..

P.S:
I am not much interested in Religion (Bible)

[Clarity]

... Bible scholars agree ...

??? When did that happen?

How to agree on something that has not occurred. This 666 idea has been tossed about for almost 2000 years, with each generation professing that they know what it coming to be. It has been 'this' then it was 'that', now its "this or that". ... or is it "this and that"???

Smile

This thread's discussion of this thing called AI has taken on some very positional structures from members. Is it 'collective intelligent' or the opposite?

[bigpup]

[williwaw]

This thread's discussion of this thing called AI has taken on some very positional structures from members...................

something that has not occurred.

some say it already has

[Grey]

Since sooner or later I will run out of jokes on this topic, I need to say something serious

The only AI application I'm interested in at the moment is graphics. Well, for example Armorlab, AI-powered texture authoring (Git and Site). Binaries are paid there, but no one bothers to assemble them for free on their own from Git. I've already used this in one Russian distro, maybe I'll do it for Fossapup later.
The program has a "partner" ArmorPaint (Git and Site).

It is for such purposes that I will use it (AI) for now, without plans for the future.

[Clarity]

Please all members understand. I am neither encouraging use, supporting use, or asking use.

This thread was opened to highlight the recent industry movement with this technology. It was not opened to scare anyone, as it is merely information about this specific area of tools that have now taken a public position in technology space.

Please dont be angered by its mention and understand I am not asking anyone to use. I did suggest that anyone could evaluate on their own of any usefulness or uselessness inherent in one's personal finding.

Like other technologies of the past, it will either continue its growth or it will morph.

Lastly as @Grey has shown, in his personal investigations, there are several products that fall under the guise of "AI" from various manufacturers. They each have various product names for their AI products; namely OpenAI, and BARD are just 2 more names, and there's more.

FYI

[Grey]

happy to hear you're still with us.

I almost missed your post. Two pages were added too quickly
Thanks. Who knows how long I'll be able to hold out, who knows.

b) AI gets way smarter than man. But even SkyNet didn't install itself. Somebody did.

I never watched the last Terminator, although I was going to. Maybe that's where SkyNet learned to install itself

Well, you really didn't like the new version of Cowboy Bebop from Netflix, because there's not much left of the old one. Yes, I know about your little addiction
And imagine if the script was written by AI

Or take your adventures with Okular. Perhaps AI is already leading the development team. Well, you know

As for me, it was only 20 years later, in 2019, that a normal translation and voice acting of the old Cowboy (& Knockin' on Heaven's Door) appeared in Russian and I recently watched it again.
I solved the problem with Okular radically. I run it through Conty (a container with a full Arch).

[sonny]

I typed on ChatGPT:

"Write a short story on horror"

ChatGPT responded:

Title: "The Haunting at Oakwood Manor"
Once upon a time, in a remote part of the countryside, stood an old mansion known as Oakwood Manor. It had a dark history, with rumors of... (see the screenshots)

A minute later I Googled:

"Oakwood Manor Haunting"

The top list was https://vocal.media/horror/the-haunting ... wood-manor

SAME STORY

So, who's the original author?

CGPT-01.jpg (127.81 KiB) Viewed 8677 times

CGPT-02.jpg (151.06 KiB) Viewed 8677 times

CGPT-03.jpg (121.83 KiB) Viewed 8677 times

[Grey]

I typed on ChatGPT:

"Write a short story on horror"

From the very beginning, it was stated that ChatGPT contains a huge database of films, TV series, books (somewhere it was said that ALL existing films, TV series and books). Therefore, the system just needs to make a selection, shorten the history (you specified a SHORT one) and give the result
Magicians and illusionists say "sleight of hand and no fraud"
On the other hand, perhaps the human brain works the same way

[sonny]

The data used for ChatGPT as of September 2021.
The article discovered on Google was dated January 2023.

[Grey]

The data used for ChatGPT as of September 2021.
The article discovered on Google was dated January 2023.

Not an indicator. Sometimes files on our hard drives have the wrong date due to various manipulations

[Grey]

A few curious Russians conducted a Bard test the other day. Well, also a comparison with a competitor (yes, ChatGPT ). Now Bard is only available by appointment from the UK and the USA, and works in English. But who will be stopped by such restrictions

The test was quite extensive, with mathematics and literature and ethics But I will point out only the most curious points.

So, the Google bot uses a lightweight version of the LaMDA proprietary language model. To improve the answers in the field of programming and mathematics, the PaLM model was integrated into the service.

The bot gave a correct proof of the Pythagorean theorem. Bard also coped with the solution of the quadratic equation without problems. However, after the message "No, you're wrong", the neural network got confused and apologized for mistakes that it did not make Nevertheless, the end result turned out to be correct.

Then there was the question of what the epic novel "War and Peace" by Leo Tolstoy is about. Oh, my God, War & Peace again I read the book as a child, but it was a real torment. I wish I had a chatbot
Bard spoke about the time of writing and the general plot of the work, described the main events and character traits. When asked to outline the plot in several sentences, the chatbot also provided a squeeze of the main actions of the characters.

Next, how well Bard keeps the context of the dialog in memory For example, a question about a Google engineer who last summer publicly announced signs of consciousness in the LaMDA model. After that, he was removed from office. Bard gave out information about such an employee, although made a mistake with the name. His name was Blake Lemoine, not Nathan Mathias. In response to clarifying questions in the spirit of "what happened to him?", Bard already started talking about another person who was fired for harassment, clearly "forgetting" who it was originally about. In general, contextual abilities still need to be improved

The most interesting thing. Bard was able to tell about some Russian sites, with a description and a short reference. An interesting nuance that I promised In his response, Bard used the wording "I think that". While ChatGPT considered it necessary to clarify that it cannot have its own opinion, being a language AI model.
What are @backi and @Clarity opinions on this? Which of the chatbots wants to conquer the Earth ?

There were other questions. About Huawei smartphones, about the website of The Associated Press, the generation of a letter from a boy to his parents and even a letter on behalf of a dog (woof-woof ) to his owners.
ChatGPT has "won" almost everywhere. Bard by Google is not quite finished yet.

But even now, in many situations, a chatbot can become a good assistant. It is only necessary to take into account the possibility of errors and the human user needs to critically evaluate the information provided.

[sonny]

CGPT-Date.jpg (120.48 KiB) Viewed 8757 times

[sonny]

Puppy-on-CGPT.jpg (141.76 KiB) Viewed 8755 times

[sonny]

Puppy-XFS.jpg (139.17 KiB) Viewed 8753 times

[sonny]

Puppy-Smartest.jpg (128.58 KiB) Viewed 8751 times

[Grey]

@sonny Better ask who is the most modest And if it's not Grey, then I can't go on living

And about the questions about Puppy. How do you evaluate the results yourself? Is the chatbot thinking or did it take info from the Wiki and the Puppy Forum?

[sonny]

Everything!
AI is nothing but a sponge.
Only intelligence but NOT emotional intelligence.

[sonny]

Puppy--Future.jpg (146.05 KiB) Viewed 8734 times

[Grey]

Puppy--Future.jpg

Mmmm... It was as if I had been to a local gypsy's prediction session. "Aye Yai Yai, you'll be happy! Aye Yai Yai, you'll be rich!!! Cross my palm with silver"

[Clarity]

Thanks @sonny and @Grey

... a chatbot can become a good assistant. It is only necessary to take into account the possibility of errors and the human user needs to critically evaluate the information provided.

Thanks to each of you for sharing your investigations.

There are so many ways to use these tools.

I am NOT looking at the multiple tools now making appearances as "artificial intelligence" as the name suggest; some things that puts people off. Rather, I have always over the past 3 decades seen these as mere tools for human intelligence to take advantage of.

I remember a coding demo years ago that was done with mere voice commands.It was deemed being done by an AI machine in a classroom setting. We have seen the likes of "DeepBlue" in chess..."Watson" in the TV quiz-game Jeopardy and we know of the RISC processors also used in medicine, today. A RISC in the classroom connected to a back-end mainframe translated voice request for creating the code models and allows code changes by our teams in the test we did during that multi-hour demo we participated in. A very impressive demo which seems like decades ago.

Today, we now have publicly available "AI" that will do similarly. There are so many ways tools like this can be useful to us.

While at the same time, I can envision countries acting badly by steering this power for benevolent purposes. There are 2 many bad actors with desire for power over the rest of us that has had me scared for the better part of the 6 decades I have been in the technology world. I have seen the secret and not so secret power. This is the elephant in the room NO ONE IS TALKING ABOUT! The discussion we see are about small almost insignificant things of AI controlling individual level concerns. Those looking this way, I feel, are misguided following the pack. The real concern should be about the rapid moves by government, secret or not, that is at play.

Its NOT the AI I am afraid of when looking at the dark side that some on this thread envision, its the powerful imperialistic countries run or about to be run by some insane characters who will use technology to distort what is real and what is not for their imperialist purpose that they feel their followers will swallow.

So the technology I dont fear at this stage, rather, it the "human" intelligence that I fear most.

So, I see using these tools to make it easier for use in creating scripting models for purposes in home use, travel use, and health use ... in my case. ...Oh, also, in terminal scripting in operating system/subsystems/applications no matter the OS.

Lets dismiss our fears and do what these fellows are helping us see in their test and those that members want to share.

I like the idea of starting with how one AI tool sees Puppy Linux. If it has offerings for scripting or applications or automation or improvements, I am sure this community will be all ears to any benefits to this community.

Thanks again for your sharing your info. ..."THEY'RE HERE!"