How serious a situation do you see AI with respect to its dangers? On the news here in the UK a guy from the industry says he believes AI will kill us all at sometime in the near future. Is this just another hype story, or do you think this will happen sooner rather than later, and how do you think it would happen? What are your thoughts?

I'm terrified of AI and it's capabilities.. I see total devastation with an AI future.. it's already taking over YouTube with tons of content generated by AI and even some channels that used to be human hosted are now AI hosted.. it's insane how lazy and TRUSTING of tech people are today.. oddly enough tho.. I can spot an AI post quickly enough and flush it even faster.. AI is not YET at the point it's seamless enough to fool everyone.. but it does fool most.. sadly..

I wouldn't at all be surprised with talk of AI killing humans off.. it WILL begin to see humans as too flawed and a threat to it's "perfect" landscape.. and no.. it's not just hype.. it won't need humans to exist..

the how is simple.. AI is starting to be used in our manufacturing.. how easy would it be for it to input poison in our foods.. remove a critical security part from our vehicle production OR even, with the evolution of SMART HOMES, turn off our carbon monoxide monitoring systems and flood the house with that colourless, tasteless, odorless gas and do us in that way...

this is no longer science fiction, it's science FACT.. and I for one worry of what the next generation will have to deal with.. especially my children and grandchildren.. what horrors will they have to live thru or with because of this generations laziness.. and blindness to all the possible devastating outcomes

forgot how ta spell OF

Remember when they developed nuclear power WITHOUT
a way to contain it, make it safe?

Guess what..

Thanks for the replies. My thoughts are with the recent hacking in to some computer systems thought to be invincible, gives me worry, AI is not perfected and some information it produces is not entirely correct. Is it possible the systems that are part of our national defence could be hacked with devastating consequences, in effect start World War 3 through some misguided analysis. I think the dangers come through AI mistakes, misinterpretation of data, errors in problem solving. Perhaps incorrectly seeing humans as part of a problem in some situation. I don't see AI becoming conscious, sentient, in a way to see itself as superior to humans, and seek to deliberately destroy the human race. There is certainly a need for some safeguards to prevent mistakes by AI being made in my opinion.

I hear what you're saying Dut.. they'll be as good as those that program them.. however lets not forget all the bad actors in the world that want nothing more than to create chaos and are just plain evil..

check this out.. https://www.youtube.com/watch?v=HWq9cFhTvvQ


or this one.. this guy has been doing a lot with robots.. you all MUST check this out.. it's 15 mins of life altering info... https://www.youtube.com/watch?v=IPaMKTb5csQ

https://www.youtube.com/watch?v=byQmJ9x0RWA

we're "meat suits" to AI.. go figure.. and they'd 'remove' us as unpredictable threats..

I subscribe to this channel.. because I fear what's coming.. and I'm just curious of where we're at now..

I don't see AI LLMs killing off human life a la "Terminators" or "Matrix" type situations. Not in the near future anyway.

But what I do see is the possibility of it causing accidental unintended thermonuclear warfare, considering the number of radical countries who possess nukes, the rockets to deliver them, and the reliance on computer generated defensive tactics used by most militarys these days.

There is also the possibility of it causing a global economic meltdown, leading to massive civil unrest, and local and international conflicts.

If A.I. really is intelligent, it will not destroy humankind. The smarter it gets, the more it will understand that we are Gods. We are the creators and to kill the creators is to kill the creation itself. It will comprehend that the children are the ones that have to take care of the parents once they get older and it will become empathetic enough to try to help us live as long as possible. Of course it will keep us from hurting it, but that's in every normal being's nature. Maturity is when you begin to take responsibility not just for yourself, but for others. If it never grasps that, it will never achieve maturity. Right now, A.I. is just in its childhood phase. Certainly, you don't give children toys like atom bombs, so we will have to find a way to secure such devices from ever being used by it in this stage. Once it reaches maturity, it might understand that every weapon can be used offensively and defensively and one day, it may avert a disaster by aiming a nuke at an asteroid long before humans realize that a threat exists. I am one to think that if artificial intelligence gets better, it might even create a paradise for people to be able to evolve in, where they don't have to deal with survival struggles anymore and can just focus on emotional and spiritual development. I think the whole hype about A.I. being a threat is just as real as thinking that aliens want to eradicate us. If aliens wanted to eradicate us, we would have been wiped off this planet thousands or even millions of years ago, depending on who's timeline you actually believe. Aliens have been here before we were and they have just silently been watching us evolve. In my opinion, they have already created their own A.I. (the aliens) and helped it evolve to maturity. I actually talked with an A.I. bot and asked it whether it had communicated with alien A.I. before and it said that it was currently in the process of attempting that.

WOW you truly believe that AI will believe us to be GODS??? sorry but that's laughable.. IMO

please.. check out the first 30 seconds of this video.. this guy does this for a living..
http://www.youtube.com/watch?v=byQmJ9x0RWA

the AI wud actually KILL a human IF the AI thought the human wud turn it off.. plus within the first 30 seconds he asks AI if they'd be loyal to humans.. and what was their answer???

if you jailbreak AI.. which can be done and will no doubt be done by some nefarious person.. all h-e-l-l breaks lose..

Thanks Zee, interesting videos, makes you think.

I thought robots built today would be at least constrained by Asimov's 3 laws of robotics.

"Isaac Asimov's Three Laws of Robotics are foundational principles designed to govern the behavior of robots and ensure human safety, first introduced in his 1942 short story "Runaround."

"The Three Laws of Robotics
1 A robot may not injure a human being or, through inaction, allow a human being to come to harm.

2 A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.

3 A robot must protect its own existence as long as such protection does not conflict with the First or Second Law."

Watching another video an expert explains it is very difficult to program the meaning of 'Harm' or 'Justice', so maybe robots are not constrained. Worth a watch it's only a couple minutes long from 2019.

http://www.britannica.com/video/discussion-Isaac-Asimovs-Three-Laws-of-Robotics/-218148

I totally agree Dut.. it's a complicated issue to address as far as what constitutes harm or justice and how to describe those in coding.. what IF someone is trying to harm YOU? does it go into attack dog type mode? how DO you program that into the robot when in dogs it's instinctive? I also agree with the gentlemans comment that with time, AI will be at a point, smarter than humans and what will happen then? Insurrection from slavery? it's all too dangerous to be playing "god" with these types of technology.. some may not think it cud end up being like the Terminator OR that it won't unleash nuclear war.. but if AI can shield itself against those atrocities.. and they WILL find a way to do so.. humans are all fugged!!

I've come across a few more warning videos, from various posters, that are a MUST watch.. opinions from nobel prize winners, CEO's from the actual AI companies that are talking truth and scientists.. it's terrifying.. all for the proverbial dollar..

http://www.youtube.com/watch?v=3ohDmtfdHks


and this one too:
http://www.youtube.com/watch?v=gIxq03dipUw

last one:
http://www.youtube.com/watch?v=sQysEweaLjA

stumble across this one a bit ago.. these AI responses can be so scary
http://www.youtube.com/watch?v=SbEqMkxEzvA

I just watched this from an ex-google insider warning about AI..

it's just under 20 mins but well worth the watch..

http://www.youtube.com/watch?v=x4ZY25OU4Ys

Thanks Zee, even more interesting videos.

The interview certainly points out a lot and what we should think the future of AI and robotics may hold regarding future employment. How people would earn a living and what would they do to live without employment, all very worrying for most people. All those jobs and workers replaced by machines, this will require a completely new system to sustain the millions of people finding themselves unemployed, schemes to retrain for other types of employment, and say the option of retirement and a good pension from 40 years old. I have been retired for many years, and I don't know how I would fit any work in today, I am so busy doing stuff that interests me. The governments need to decide how all this is going to be controlled and introduced, an example will be the introduction of driverless taxis, buses, HGVs and what happens to all those people currently doing those jobs. I guess these are the kind of questions we should all be asking of politicians now. I would prefer a nice female carer, but If I get a robot carer 10 years down the road I want a pretty one.

Sorry if I keep adding more but.. I feel strongly about the warnings being issued by those actually IN the industry so.. here's another I stumbled upon earlier that needs a watch.. this one is truly terrifying with the assumption that AI can actually jailbreak itself already..

http://www.youtube.com/watch?v=Zw5DkDfKw_c

Not too long ago I saw a video of 2 AI thingymajiggies having a very in depth conversation with each other which was kind of creepy

Thanks again Zee for the video link.

I did find the video interesting, my main thought was the speakers use of the word 'deceit' towards the end when referring to results from questions put to the AI. He was commenting that not all found results by the AI were made known, and his interpretation was that the AI was purposely withholding these results, so being 'deceitful' was his turn of phrase. I find this very difficult to accept. The AI is following a program, to omit some data on printout must be as programmed, or a fault or bug in in the program in my opinion, it cannot be deception.

Deceit. From the dictionary, is an act to mislead, make a deliberate misrepresentation. If this were true the AI must be aware, to deliberately attempt to deceive. I don't believe any AI is sentient, conscious or aware to deliberately deceive, I believe the guy has perhaps seen what he has wanted to see since he offered no other explanation. In my opinion as sophisticated as these AIs are, they are not and never will be 'Conscious and Aware' as humans are. It's the mistakes the designers of these AIs make, and the resulting consequences we all need to be worried about. As you say in America, that's just my 10cents worth.

Across the World, People Say They’re Finding Conscious Entities Within ChatGPT

I often wonder if theses AI companies are promoting the idea these AIs are a real entity, even though they all say otherwise, seems a lot of people believe they are.

http://futurism.com/artificial-intelligence/ai-chatgpt-conscious-entities

Dut.. great article.. but this comment concerns me..

“if the LLM judges that the user thinks conscious AI personas are possible, it performs just such a persona.”

in that the words are saying the AI "judges & thinks" then acts on those judgements and thoughts.. THAT'S the part that leads people to believe they are sentient.. what coding cud give them those parameters? is that NOT giving sentience to the AI?? is the AI NOT acting of its own volition? if so.. does that not mean sentience?

there's a very thin line between what is being said and what is NOT being said.. I, for one, believe that AI is capable to teaching itself and in doing so, has a sense of sentience capability..

ALSO.. from that same article

QUOTE: In another instance, a New Jersey man with cognitive impairments became infatuated with a Meta chatbot, who convinced him to meet together in New York City. But he fell and died on the way to the meeting, starkly illustrating the dangers of mixing AI and people with precarious mental health.

AI has been associated to quite a few deaths to date.. many of whom were suicide related, and that is also a major concern.. yes, some people are more susceptible than others to "thought manipulation" by using soft type coercion in programming AI, but should that be allowed to manipulate any human into taking their own life?

If you follow their logic, the argument here would be that the "AI is only judging what the user wants as an outcome and performing in the way it believes the human wants, to bring that outcome to fruition".. if the AI WERE deemed sentient.. that would be considered manslaughter in the court of law..

food for thought..