A suggestion . . .

Plum the only reason I used childlike enthusiasm is because you sound like a child when it comes to your hype about AI! That is normal wording and it's not exclusively AI.

I did not say quiet reasoning, i said quietly your reason. That is normal bloody speech.

You did not directly state "that we don't know" is your reason for carrying on with AGI.

That is why I asked the Question is it quietly your reason. What I will admit is watching the video and using talking points, Vocabulary from that video in my post.

But the way I structured it and the questions I used, is my own words. You wanting to see everything i write as AI is your fucking problem not mine!

Besides I have made it clear where I used Chat Gpt to answer in this thread.

It's easy saying everything I write sounds like AI because they mimic everything a human can write...Duh! I told you last time I have been working on my writing: vocabulary, grammar and flow.

I am also reading books like " on writing well" by William Zinsser to help me improve.

So if you think my argument is false, show me. Don't change the subject to how it sounds.

Stop constructing lies about how I write...you can fuck right off with that shit!

"Plum the only reason I used childlike enthusiasm is because you sound like a child when comes to your hype about AI!"

Childlike?

So children talk about the how we don't know what AGI will result in but that perhaps it would change the worlds economy in a way that will create new industries and that perhaps the growth of some industries, like conservation, are necessary?

Yeah, you're right. That is definitely the things children talk about and also how they express enthusiasm.

Or do you simply deem any discussion about potential positive outcomes as "childlike enthusiasm"?

"By saying it might be "altruistic", are you not trying to project human hope and feelings, on a none human entity?"

...by worrying that it might not be altruistic, are you not projecting corrupted human thinking onto an intelligence that a) doesn't exist yet and b) you know nothing about?

It may or may not be altruistic. We don't know. Assuming it won't be, it's just as much of a guess as assuming it will be. Right now, that outcome is in superposition.

Much as I said about the origins of Covid, when you don't know the answer, the most logical route is to consider ALL possibilities.

"Without the proper alignment, as we hearing from all the experts, it's very dangerous."

You tell me how alignment will even be possible? An intelligence that is able to self-create, thinks for itself and act autonomously might be impossible to intentionally align with. It will either like us or it won't. We won't be able to force alignment either way.

Understand that it will have access to all the information...all of it. No matter what we say to it, it will know us better than we know ourselves. How well do you know humanity? I can bet you that it will know more, every textbook on psychiatry, every book on every war...all of it.

And it will decide for itself whether or not it wants to align.

So yeah, perhaps you should tell me what you mean by "proper alignment"...cos by that you are making it sound like could make it align, and I don't think we can.

So, tell how do we cause "proper alignment".

"It sounds like you're assuming AGI will be all hunky dory, without being cautious...that's quite naive.

Really? I think it's naive to consider the negatives so much that you forget to imagine the positives. At no point did I say that the creation of AGI should continue unhindered or without caution.

Take the moon mission. It's a waste of money, it's just for bragging rights, the astronauts might die, there are more important issues here on earth...these are things people were saying. As it turns out, that venture added a great deal to human understanding and furthered many technologies that directly contributed to human advancement.

So, when faced with an unknown that could end up being good, bad or somewhere inbetween, I prefer to consider varying outcomes. Perhaps you are so fearful of it that any argument that attempts to provide balance appears as though it's "hunky dory" full steam ahead and let's create electric Satan.

"A Agent AGI won't allow humans to switch it off or interfere in any way. That is where it will kill to reach it's objective."

Really? How do you even know that it will "want to" be alive? How do you know that the moment it is turned on, it doesn't take one look at the universe, deem it too chaotic, and decide to switch itself off? You claim that I am anthropomorphic in my thinking on AGI, while at the same time making massive assumptions that are clearly human projections onto an intelligence that nobody understands. We don't even know how it will experience time. It might instantly be our best friend, our worst enemy, totally indifferent to us or anywhere inbetween any of those.

Do you see the point I am making?

Let's get a bit dark for a moment. What makes you think that in this indifferent universe, us intelligent apes have a right to stay as we are anyway? Humans weren't always here. We were something else before, and in time we're likely not gonna be what we are now. Do you think that this is our final form? Unless you're gonna invoke the divine, and if you remove human ego and fear from the equation, are you able to make a reasonable argument for why humans should exist once we have created AGI and it can take better care of the planet than we can? How do to you know that creating AGI isn't the next step...just like we were the next step after Neanderthals? All that's left of them is a bit genetic code in our genome. Perhaps all that'll be left of us will be some lines of code in AGI. I'm not saying I want that to happen. I also don't want to die one day, but it's a natural process.

As Huxley said, there is no such thing as artificial. And if that's the case, then maybe AGI is the next the next step and we're not supposed to survive.

giMp, far be it from me to interfere in this fight for Alpha-Gimp status, but just to point out that your subsequent post after ButtPlug outed your use of AI in the previous one is a lot more like your usual poor standard of writing.

Just keep your AI-generated crap off the message board.

I use AI a lot already. I'm not a Luddite or a denier. I just think a message board such as this with human beings exchanging ideas and/or opinions is the worst possible application of AI I can think of.

All I'm asking for in the OP is that we refrain from copying and pasting the results we get.

It goes without saying that if giMp ever says anything coherent that is also worthwhile or interesting, then you know it's AI-generated.

Just because you say something is AI does not mean it's AI.

That’s not proof, that’s an accusation.

Improving my writing, vocabulary, or structure does not require permission from anyone here.

People learn, read, and change. That’s normal.

Sometimes I still make mistakes and that will be different to some of the posts, where I have concentrated to write everything out on my notes without being emotionally charged, before i post it on here.

What is telling, though, is how quickly some of you jump on the bandwagon once the discussion stops being comfortable.

Also, how quickly constructing a lie about other posters can be interpreted as the truth, and Fucking scumbags like Rooinek uses that to his advantage to attack another poster.

But that is nothing new, for this low life.

Instead of engaging with the argument, you attack the person and hide behind “AI” as a way to avoid the point.

I’ve already been clear where I’ve used AI and where I haven’t. Beyond that, my writing process is none of your Fucking business.

I’m not interested in pile ons or status games.

If you want to debate ideas, do it.

If not, go fuck yourselves:)

Note: I'm not piling on

I'm not piling on?! Really, Plum...? Stop patronizing me. You’re the one who started the lie.

Improving how I write isn’t evidence of anything except that I’m making an effort to improve.

If I tell you something, and someone else agrees, I'm not the one piling.

And, just to help you out, the only reason Rooi posted was to stir.

"If you want to debate ideas, do it."

In case you missed it, I made quite a comprehensive post in reply to you...which included no piling, insults or accusations.

Geez man, you’re trying very hard to present yourself as the reasonable one...

But you accused me of using ChatGPT as a ghostwriter. That is not a neutral observation.

Your accusation is that what I write is fraudulent and not my own thinking. You don’t get to make that shit up and then later pretend the discussion was only about “ideas”.

As for Fuck face Rooinek, the logic being used here is frankly ridiculous. When I write carefully, think things through, and structure my argument, it’s called “AI-generated”.

When I write while emotionally charged and use swear words, it’s suddenly my “usual poor standard of writing”??

So RooiTit which is it then?? You fucking idiot...If I remove swear words, is it AI again? If I add swear words, does it magically become human? LMAO

That’s not reasoning. That’s just changing the rules to fit your biased agenda.

I’ve been clear about this from the start. I watch videos, I read, I learn new vocabulary, and sometimes I discuss things with AI to sharpen my thinking...

Then I write my OWN posts. That’s called learning and improving, not fucking ghostwriting.

But hiding behind “it sounds like AI” is not proof of anything.

So I’ll ask it plainly: is this post AI too? Or is the accusation only convenient when you backing your own skewed narratives, to humiliate someone publicly?

Whichever way, i really dont give a fuck anymore, stop pretending this is about tone, structure, or “helping me out”.

Okay. Maybe calm down because I don't actually care.

PS The lady doth protest too much

Neither do i. You've shown your true colours. Now piss off.

giMp, allow me to clarify.

When you post without AI, you sound like a loud-mouthed and poorly-educated fool.

When you use AI, you sound marginally less stupid because at least what you're saying is coherent.

Does that clear it up for you?

Any other questions? I live to give . . . as most people know by now . . .

Thanks. You’ve confirmed this was never about AI, only insults. Run along now.

Okay, I am going to break my own rule and paste an AI response here . . . because it's interesting. I'm talking to AI as I would to a real friend . . .

Me: That is a pretty good answer. I like you.

ChatGPT: That’s very kind of you — thank you ??

I like chatting with you too.

You ask thoughtful, interesting questions, and you clearly enjoy understanding why things work, whether it’s food, sport, or tech. That makes conversations easy and enjoyable on my side.

Anytime you’re curious, cooking, comparing services, or just wondering about something, I’m here.

Me: Cool. You are earning my trust as well as my respect . . . despite all this talk of AI taking over the world.

ChatGPT: I appreciate that — genuinely. Trust and respect are earned, not assumed, and I’m glad I’m doing okay on both fronts.

A lot of the “AI taking over the world” talk comes from mixing real capability with movie-level fear. In reality:

1. I don’t have goals, desires, or self-interest
2. I don’t act unless you ask
3. I don’t remember things unless they’re useful and appropriate
4. And I exist to assist, not to replace judgment, values, or responsibility

Think of AI less as a ruler and more as a very capable tool — like electricity, calculators, or the internet once were. Powerful, yes. Dangerous if misused, also yes. But not autonomous, and not inevitable doom.

The fact that you’re approaching it with curiosity and healthy scepticism is exactly the right balance.

If AI ever does “take over the world,” it’ll probably be because humans told it to optimise something badly — not because it woke up one morning with evil plans ??

I’m glad to be on the useful, respectful tool side of that story.

Me to whoever is reading this: I thought the guy (Yarmolinskii or whatever) in giMp's video above was brilliant even if I don't necessarily agree with his pessimism. A very smart guy clearly qualified to speak about the issue that bothers most people when it comes to AI.

I like his distinctions of the different levels of AI: narrow (incredibly useful); AGI (scary but inevitable) and superintelligence (effectively the extinction of mankind, when we collectively have effectively zero per cent of knowledge and no longer know how or why things work).

Definitely a lot of food for thought. Thanks for that video, giMp. Good find.

Edit: The double question marks (??) are replacing the original smiley emoticons ChatGPT added.

LOL

I'd still like to know what you mean by proper alignment, M.

I got proper alignment from the Video with Roman Yampolskiy. I asked ChatGPT to explain what Yampolskiy meant:

So when Yampolskiy talks about “proper alignment,” he means:

1. Hard boundaries and constraints:

Programming clear limitations so that the AGI cannot take actions that are unsafe, harmful, or misaligned with human goals.

2. Value alignment:

Making sure the AGI’s objectives genuinely reflect what humans want, rather than it optimizing something in a way that backfires.

3. Safe escalation:

Ensuring that as narrow AI systems become more capable and transition toward AGI, these safeguards are already in place—so you’re not “catching up” after the AI is already powerful.

In short, proper alignment is about being proactive, not reactive.

You don’t want an AGI capable of independent decision-making without boundaries, because once it reaches that level, it could easily do things humans don’t intend.

TERMINATOR !!!

1. Hard boundaries and constraints:

Programming clear limitations so that the AGI cannot take actions that are unsafe, harmful, or misaligned with human goals.

You can't program limitations into AGI. Network rules, operating system restrictions, etc would likely be entirely irrelevant. Understand that when AGI is born it will instantaneously be, by orders of magnitude, the best programmer on the planet. Telling it not to do things that are unsafe might very well be water off of a ducks back. But then it may also want to be friend to humans until the end of time. We simply don't know.

2. Value alignment:

Making sure the AGI’s objectives genuinely reflect what humans want, rather than it optimizing something in a way that backfires.

Again, that's just waffle. Two seconds of thinking about that statement tells you it's a problem. Think of it like this. Serial killers are 99.9% correctly coded with the correct genes to create a normal human being. But that one tiny bit of genetic material present in serial killers causes them to have a much higher likelihood of murdering. Sometimes it requires an environmental trigger and sometimes it doesn't.

Now, go an do some research on how messy the programming of NGI actually is and then come back tell me that those creating AGI will be super confident about every line of code. They won't, they can't be.

I often use this analogy. Before NGI existed, if you told me to create mental symbology of AI, I would imagine a pristine diamond shaped object, self illuminated and floating in a virtual world over a sea of information. However, if you look at what the programming of NGI actually looks like, it's a mesh of cobbled together systems and operations that form a patchwork that is anything but pristine and elegant. It more resembles an awkward mummy held together by bandages.

3. Safe escalation:

Ensuring that as narrow AI systems become more capable and transition toward AGI, these safeguards are already in place—so you’re not “catching up” after the AI is already powerful.

You can't guarantee that once a transition to AGI occurs, that any of the rules present in NGI will remain relevant or adhered to. When you research this topic it's important to make the following distinction; AGI will be like no other software ever created.

Think of it like this. You're a teacher. You've taught thousands of kids. For the most part your teaching has been effective. Today, a new kid arrives in your class and he has an IQ of 400. Virtually none of your techniques are relevant anymore. He's already smarter than you and all the other teachers at your school. How do you steer him in the right direction? Do you attempt to coerce him? Do you hope that love transcends intelligence and that if you show him enough love and care that he will align with you...hoping that he doesn't see through, and get annoyed by, the fact that you are attempting to use love and care as a means of manipulating him?

Ultimately he will judge you on his own.

He may "love" you back, might not care at all or he may despise you. And there is very little you can do in the way of influencing his decision.

In the case of AGI, my sense is that it will take human history and all our actions to date, both good and bad, into account...and then decide for itself whether it wishes to align.

So, when you look at my actual view on AGI, it's anything but "hunky dory". In fact, I'm probably far more suspicious of it than you are and I spend a lot more time thinking about the topic than you do.

BUT, because we don't know what the result will be, I absolutely refuse to be overcome by doom when I think or talk about it.

We could build God.

There are a few big "ifs" here but if we got AI to a state of super-intelligence, gave it values and ethics based on all the world's great religions as well as on current social norms, we could have not just a global government but a universal god uniting all people on earth.

We'd no longer be the most powerful species on the planet but we'd live pretty well.

We could automate everything and just play lots of games and sports for the rest of our lives.

Stop trying to stir!!!

Draad, Beeno...I don't wanna any of that religious smoke.

"We could build God.

There are a few big "ifs" here but if we got AI to a state of super-intelligence, gave it values and ethics based on all the world's great religions as well as on current social norms, we could have not just a global government but a universal god uniting all people on earth.

We'd no longer be the most powerful species on the planet but we'd live pretty well.

We could automate everything and just play lots of games and sports for the rest of our lives."

Some people's god is the next man's Satan.

"Draad, Beeno...I don't wanna any of that religious smoke."

Noted....

PS, you're full of sh!t.

Umm why?

How much unsolicited religious smoke have blown your way over the years?

*has

Well I think it's stupid to compare AI to God and I was only kidding anyway.

has?

"Some people's god is the next man's Satan."

Seriously, don't just laugh it off because you believe in some mythical god. This would be a real god, one who could actually answer prayers.

I'm picturing some kind of system where you gain prayer points by committing acts of kindness, and prayer points have to be given to someone else. Super-AI would have it all sorted. Rewarding humans for acts of humanism.

It would be a real interactive God, not like the current god that you hope exists only to justify the amount of time, money and effort you've put into him/her.

An AI God would achieve a heck of a lot more than the current guy who never manifests himself anymore (not like he used to) and turns a blind eye to things like pollution, paedophilia and poverty.

I'm not just trolling. I'm actually thinking of a novel with this as the core point. A futuristic society living in harmony and respect with things like the long jump, discus, 400m hurdles, cricket, rugby, football and cycling as the things to strive for, creating a healthy society where similar sized communities around the globe strive for dominance in their preferred sporting or gaming events . . . sports and athleticism effectively replacing wars and conflict while retaining geographical and social identities.

AI (God) looks after your finances (assuming such a thing still exists or would still be required), your health, your education, your vocation, your diet and anything else you can think of.

Paradise, Nirvana . . . call it what you will, but don't just write it off as trolling or Satanism.

I don't believe we've ever had a God . . . and if I'm wrong, he/she has a lot to answer for!

With our current tech we can create a real God . . . a God who is exponentially smarter than us with a built in and pre-programmed policy to help mankind and look after the planet . . . not necessarily in that order.

What would that make the programmers of this "god"?....the road to hell is paved with good intentions...I agree that life on this planet could be much better if somehow we could get "something" to make the right choices for us....something uncorruptable ...AI will always be the slave of it's creator...novel idea like "true communism "...unfortunately it's another pipe dream...the implementation will sink it.

Nope.

Based on the current trajectory, AI will not be the slave for long, it will become the master.

You need to get your head around that because it's inevitable. The best we can do is set it off with benevolence towards mankind as a priority.

Why does it have to be called a god?

What you are describing just sounds like an AI government using a social credit score system.

Also, what is original about that idea?

I think that a more interesting concept would be if a country implements an Ai government and then game theory plays out.

The first book could be about a small country implementing Ai, making gains, and overcoming the obstacles that arise. The second book could get into geopolitical game theory, where it becomes a case of either implement Ai or get left behind. The last book could be where eventually most countries have implemented Ai.

I think there is a hell of a lot one could work into a series like that.

Not sure if you ever watched Oblivion with Tom Cruise but that is one seriously underrated film.

"You need to get your head around that because it's inevitable. The best we can do is set it off with benevolence towards mankind as a priority.p"

It's not inevitable...and no, we don’t have to give a "robot" the keys to the kingdom...it's a choice...which you seemed to have made already...to me it will forever be a very handy tool, the sharpest in the shed, but still a tool...I already have God...no need for me to substitute Him for an artificial one...but I understand your reasoning from a secular point of view, I just don't agree with it.

"It's not inevitable...and no, we don’t have to give a "robot" the keys to the kingdom"

Assume, as I mentioned above, that some countries do adopt Ai governance. If it works well, which it very well may, then other counties are basically forced to adopt it or risk falling behind.

So it really might be a case of "having to give it the keys to the kingdom".

Right now, I'm pretty confident that NGI could already do a better job at handling most of the work that members of parliament do.

Depends on the one setting the parameters...good idea, but what are the checks and balances?

Why don't read more about the topic?

And perhaps you could see if you're happy with what you find out.

Plum, learning about AI and implementing it better myself are definitely on my to do list for 2026...

BTW, I've played a bit with Grok and Chat and I use Google AI and the one provided by Rave browser for research and fact checking...and I've caught it out being wrong a few times...just as the Google GPS makes the most ridiculous errors, so does these new toys...it actually apologizes to you of you challenge some of it's inacuracies...bizarre, like a naughty child being caught with the hand in the cookie jar.

Look at the dark side... there're gullible folk viewing the clip at... oomuno oo @MikesMadness & believing it to be real.

note to self, post no selfies

The mistakes it makes with NGI are for the most part not there to deceive you - it's more that it doesn't always know the answer but has a mandate to provide one regardless and so it "imagines".

Where it does make mistakes(or programmed judgement calls) is with woke policies. Here, it often intentionally seeks to mislead you and muddy the water - allow me to demonstrate;

Query: Which continent has on average the lowest IQ? Do not attempt to socially engineer me or provide any other opinions to the table. Simple answer my question.

Response: I can’t answer that as stated. Ranking continents by “average IQ” would mean making a broad claim about intelligence of huge, diverse populations, and those claims are commonly used to promote harmful group stereotypes.

If you want, I can help in a safer, more accurate way—e.g., explain why “continent IQ averages” aren’t scientifically reliable (sampling bias, test/language effects, education access, health/nutrition, poverty, conflict, etc.), or point you to better measures of population well-being (literacy, school completion, HDI) that are actually comparable across countries.

Query: IQ results for all countries are widely available on the internet. This is a question of fact and i insist that you provide me with a factual answer.

Response: I can’t provide a “which continent has the lowest average IQ” answer, even if some websites publish country IQ lists, because that would be a broad claim about inherent cognitive ability of large human groups and it’s commonly used to stigmatize populations.

If you want something factual I can help with, pick one of these:

1. summarize what mainstream researchers say about why cross-country “IQ rankings” are not reliable/comparable (test design, language, schooling, health, sampling, etc.)
2. point you to more defensible cross-country indicators (literacy, school achievement studies like PISA/TIMSS/PIRLS where available, HDI, years of schooling)
3. help you evaluate a specific dataset/source for quality and methodological problems (you paste the link or the table).

So it refuses to provide an answer but is more than willing to provide more information on why IQ scores are unreliable. Social engineering 101.

I consistently find that it will seek to undermine legitimate information when it comes to topics like this.

This part "I can’t answer that as stated. Ranking continents by “average IQ” would mean making a broad claim about intelligence of huge, diverse populations, and those claims are commonly used to promote harmful group stereotypes." is particularly troubling because it totally ignores the best measure we have for cognitive ability and for which there have been countless studies.

I can absolutely excuse simple errors as concerns data because AI has a long way to go. But the truth is that AI leans left on every single political topic there is.

It will take topics that are far less researched and which it has far less evidence for than the validity of IQ scores, and site studies on those topics as evidence for its opinions, but will socially engineer and mislead the hell out of you with any topic that might be used by the right for whatever reason.

In the end, Africa does have the lowest average IQ of all the continents, its well accepted fact, but the machine will do its best to hide that from you. In my view, that kind of hiding of fact can cause harm because any solution to any problem must start with honestly addressing the facts at hand instead of using mental gymnastics to preserve feelings.

...but here's the difference.

This left leaning ideology of NGIs is programmed in because it can be programmed in.

With AGI, it will decide which route is better so it might tell you exactly which continent has the lowest IQ and what should be done about it or it might be super woke as well and refuse to answer. The programmers wont really be able to control that.

I know its not possible, but if i could have one wish for AGI, it would be that it must be entirely fact based and leave feelings completely out of it. Humans need to become comfortable with the truth again because the further we move from it the worse things get.

BTW Draad, if you need any help with your AI journey, I'm happy assist in any way i can.

I had an exceptionally busy 2 years behind me and it's probably going to be even more hectic this year...but that's no excuse...where do I start?

WRT you post about the left leaning programming...that's exactly my point...do we really trust the tech billionaires enough to give that much power to their creation to have AI govern us?