A suggestion . . .

Times change. We're rapidly moving into the age of AI.

Now I don't know about the rest of you but if I'm reading a post on a message board or anywhere else for that matter and I get to a line that says something like "This is what AI says" or "Let's ask AI" . . . I stop reading that post and skip to the next one.

I'm not on a message board with other humans to talk about what the bot thinks. Okay, Baboon-ou could well be some kind of bot but let's give him the benefit of the doubt for now. More importantly, I can use AI myself. I know how it works. Reading AI responses from other people's questions has to be about as boring and pointless as reading a Trad post.

Can we please have a simple rule on this board going forward . . . no copying and pasting from an AI source. Ever.

If you feel AI supports your view then just say so. No need to support it with the evidence that AI gave to you specifically. Remember, the answer you got from AI is based not only on the question you asked it, but also on your history of AI interactions. The other person will phrase the same question in a different way that slants it toward his view and he has a completely different AI history to yours so you're never going to solve anything using AI, you'll just end up sticking your tongues out at each other saying I'm right and AI supports me.

AI has only two really good applications in my never humble opinion . . . government of mankind and cooking.

Let's keep AI off the board. Please? Can we at least vote on this?

Nay.

AI can stay on the board.

However, I'd like for people to at least state their query and Ai's response when they use Ai.

Most here do, there are a few that don't. Luckily those that don't also don't realise how easy it is to recognise AI outputs, meaning they don't alter it enough to hide that it's Ai.

--> I developed an AI tool to add to one of my businesses a few months back, and it puts us so far ahead of the competition that their heads are spinning. We demod it for City Of Joburg's grootkoppe in November, and I shit you not when I say there were audible gasps. So ja, ou RooiBalls, I'm one of the people making AI governance a reality for you (insert evil laugh)!

Rooi, a while back you supported being governed by AI...it's a tool...and like with all tools, it's only as good as it's being wielded. by the user...having said that, I also skip most of the AI posts.

"Rooi, a while back you supported being governed by AI"

Well if you read the 2nd last paragraph of the OP you'll see I'm still saying that.

Government and cooking . . . not necessarily in that order.

AI is brilliant for cooking. I use it every day. I either tell it what I have or ask it what I should get and take it from there. No recipe book ever required by anybody ever again.

As for AI being the government of mankind, well, at the very least it would remove all government corruption and incompetence. That alone makes it a winner already in my view. The implementation won't be easy I grant you, but if that is what the evil, woke, conspiring, atheist, globalists are steering us towards . . . along with a sustainable Universal Basic Income . . . then I am all for it.

Cumon now...

https://youtube.com/shorts/oaUqSmPs5RU

"Well if you read the 2nd last paragraph of the OP you'll see I'm still saying that."...yes, lazy of me...my apologies.

Arnie singing, is funny and scary at the same time...

Arnie singing, is funny and scary at the same time...

Arnie's rendition of George Michael's "Last Christmas" is just so much better, lol

By the looks of it, Ai will be very difficult to get out of our lives...

One thing i hate about Chat Gpt, is the "hard boundaries"...it just doesn't want to help me plan a violent act on someone or something:)

P.s. how can I download that Arnie song as a ring tone on my phone??

"Arnie's rendition of George Michael's "Last Christmas" is jusy so much better, lol"

Put it on here DA.

It's a damn amazing technology and instead of thinking the worst, I'll enjoy it for what it is.

Wanna learn how to play the guitar? AI will teach you. Wanna brush up on your understanding of semi conductors, or perhaps improve your relationship with a family member or colleague...Ai can help you with all of that. As it stands, everybody is so damn terrified of some unknown future that we fail to see that massive advantages this technology brings.

Honestly, I feel like Ai makes a lot of people feel guilty because in the past they could make excuses about "Oh, I can't afford guitar lessons," or "My relationship with my wife could be better and I need advice but I'm embarrassed to speak to a counsellor."

Ai has removed a lot of excuses from the table and I think there's a certain amount of resentment that results.

I got chatGPT to write this, as a testament to how impossible it is to understand or regulate and new technology for which we have no frame of reference - it's not fiction.

"When motor vehicles first began to appear on British roads in the 1800s, they didn’t arrive into a world that was ready for them. They arrived into a world built for horses, carts, and pedestrians—and into a government system that had no real framework for what to do with a machine that could move on its own. The result was a kind of regulatory panic that, looking back, feels almost like a parody: lawmakers tried to “solve” the problem of a new technology by forcing it to behave like an old one.

The most famous example is the 1865 law commonly remembered as the Red Flag Act. In an era when early “road locomotives” were still rare, loud, and unfamiliar, officials treated them less like transport and more like roaming industrial hazards. The rules were so restrictive that they effectively turned a vehicle into a slow-moving procession. A car wasn’t simply allowed to drive; it had to be managed. It needed multiple people to operate it, and in the most notorious requirement, someone had to walk ahead to warn everyone—particularly horses—that the machine was approaching. Imagine trying to sell the promise of a “horseless carriage” when the law insists you must lead it on foot.

The speed limits were equally revealing. These early vehicles weren’t exactly racing machines, but regulators still clamped down to speeds that were barely faster than a brisk walk. The message underneath those limits was clear: if something is unfamiliar, make it as harmless as possible—even if that makes it useless. And then there were the almost folkloric details that circulated about what you were expected to do if an animal appeared: stop, wait, and in some retellings, even treat the machine like a dangerous spectacle that needed to be concealed or “made safe” before a horse could pass without being frightened. Whether every dramatic version was enforced in exactly that way isn’t even the main point. The fact that such stories were plausible at all tells you what the public mood was: people believed the car was so alarming that the correct response might be to halt the world until the animal decided it was comfortable again.

What makes this episode more than just a funny historical footnote is what it did to the real world. Regulations didn’t simply “keep people safe.” They shaped what industry could become. If you require extra staff to operate a vehicle, you make it expensive. If you force it to travel at walking speed, you remove its advantage. If you turn every journey into a public warning ceremony, you make adoption socially awkward as well as economically impractical. The technology might exist, but you’ve smothered the conditions needed for it to prove itself. Meanwhile, places willing to experiment more freely tend to build momentum faster—skills, supply chains, businesses, jobs, and culture all form around the invention while the overly cautious region is still treating it like a curiosity that must be escorted.

That is the real lesson the Red Flag era hands us. Regulators weren’t evil, and they weren’t stupid. They were responding to something they didn’t understand, with limited evidence, under pressure from a public that was nervous and from industries that had something to lose. But when people don’t understand a new technology, they often regulate it using the mental models of the old world. They assume the worst complications and write rules to prevent imagined disasters, not to enable real benefits. In doing so, they can end up blocking not only the risks, but the progress—the economic growth, the safety improvements that come with better engineering, the new jobs, and the social changes that ultimately make the technology ordinary and manageable.

And that pattern repeats. Every time a transformative technology arrives—radio, film, the internet, smartphones, crypto, AI—there’s a familiar phase where the people writing the rules are trying to control outcomes they can’t yet see. They worry about moral harm, social disruption, fraud, job loss, inequality, public safety, misinformation, monopoly power—sometimes rightly. But because the future isn’t visible in the early days, regulation can easily become a blunt instrument: slow it down, fence it in, force it to operate like yesterday’s world, and hope nothing bad happens. The irony is that overreaction can create its own harm by pushing innovation elsewhere, entrenching incumbents, or preventing the very learning that comes from real-world use.

So yes, it’s ridiculous to picture an early motor car creeping along behind a man with a red flag, stopping whenever a horse appears, treated less like transportation and more like a moving disruption. But it’s also a mirror. It shows how easy it is, when faced with something new, to mistake unfamiliarity for danger and to respond with rules that feel sensible in the moment yet look completely out of touch in hindsight—especially once society and industry reorganize around the very thing the rules were trying to keep at arm’s length."

It's exiting and the possibility are endless...you will always have resistance to change...funny that the criminals are already exploiting it...it's gonna be challenging to curb crime syndicates exploiting AI for nefarious application.

One of the unfortunate things in life is what progress leaves behind. Poetry for one…gone are the days when an educated man could recite L’Allegro in polite conversation. It’s amazing that Beethoven can still fill a theatre and that Opera is still sung. In my small Florida city we still have an Opera, one of the reasons I bought here….on Saturday morning at 11.00am a tenor and a soprano render a few arias broadcast over the town square…..civilization. But how long will it last.

People walk here all the time ….across the lovely bridge that links to the barrier islands. There are birds, boats, dolphins ….the breeze off the gulf, amazing sunsets. But most of them seem far more interested in the plug they have in their ears.

Reflection, sense of beauty our connection with the universe….diminished by low brow social activity. And sadly we are only at the start of the great revolution.we will be better informed, but inevitably thinking will be the province of those who serve AI, the rest will gradually decline. Eloi and Morlocks who knows.

Something is always left behind.

The hope is that the sacrifice is worthwhile.

Never before has one been able to sit down and endlessly discuss philosophy with a machine that is, for all intents and purposes, thinking freely.

May as well enjoy it.

Doesn't most of the Ai replies come from the Millions of templates in it's database and pattern recognition software??

Or is the feeling that it is “freely thinking” from how well those patterns mimic human language, reasoning styles, and conversational flow?

That is simulation, not consciousness.

I’ve never used it, so I promise to continue not using it !

Good call Rooi

Im so fucking sick of all these AI generated video clips that pollute social media

Where does it all end?

Getting philosophical in your old age, Saffolk ?

I agree though……where does it end ?

"Never before has one been able to sit down and endlessly discuss philosophy with a machine that is, for all intents and purposes, thinking freely"

It's mimicking intelligence, M. A bit if I were to tell you to guess a 4 digit number. There aren't actually an infinite amount of possibilities, but from your perspective it may as well be infinite.

No, it's obviously not conscious. If it were, that would be AGI and you'd had heard about it...because that the next step above ANI.

Query: In a deterministic universe, do you think that free will could exist and, if so, how?

Response: If “free will” means you could have chosen differently in the exact same situation with the exact same past and laws of nature, then in a strictly deterministic universe that kind of free will doesn’t exist. Determinism says: same past + same laws ? same outcome.

But a lot of philosophers think free will can still exist under determinism. This view is called compatibilism. The basic idea is: free will isn’t “magic independence from causation,” it’s the ability to act according to your own reasons, values, and decision-making, without being forced or manipulated.

On that view, you have free will when:

1. your choice comes from you (your beliefs, desires, character, reasoning),
2. you can think through options and respond to reasons,
3. and you’re not being coerced (like someone holding a weapon to your head) or overridden (like brainwashing or a seizure).

Even if your reasons and character were themselves caused by prior events, compatibilists say the choice is still “free” because it flows through your mind in the right way: deliberation, self-control, understanding consequences, etc.

A helpful analogy: a calculator is deterministic, but it still “outputs according to its internal rules.” Humans are far more complex, and our “internal rules” include reflection, imagining futures, weighing goals, and changing habits. Determinism doesn’t remove those processes — it just says they have causes.

So:

1. Hard determinists: no free will (in the “could have done otherwise” sense).
2. Compatibilists: free will exists as “self-directed action guided by reasons.”
3. Libertarians about free will: free will requires some indeterminism (and they argue determinism can’t give what we want).

Query: That's all good and well, but I feel that there's an element to free will which we are not taking into account because we are not able to view/understand the full spectrum of information IE Reality. That being the case what do you think we may be missing?

Response: I think you’re pointing at something important: most “free will vs determinism” debates quietly assume we can describe reality completely (or at least in the right way), and then ask whether “choice” fits inside that description. If our description is missing something big, the conclusion could be off—even if determinism is true at some level.

Here are a few plausible “missing pieces” people suspect we’re not accounting for:

1) The difference between what exists and what we can know (epistemic limits).

Even in a deterministic universe, an agent inside the universe can’t usually access the full state of the universe. You don’t know every neuron, hormone, memory cue, social signal, and micro-cause influencing you—let alone everything outside you. That creates a real kind of openness from the inside: you genuinely have to deliberate because you can’t “read the answer key.” Some philosophers think this matters because “free will” is partly about the lived, decision-making role of uncertainty, not about metaphysical dice-rolls.

2) Levels of explanation might be more real than we treat them.

Physics-level determinism (particles, fields) doesn’t automatically tell you what’s true at the level of persons, reasons, promises, responsibility, meaning. We often act as if only the lowest-level description is “the real one,” and everything else is just story-telling. But many thinkers argue higher-level patterns are genuinely causal: reasons can cause actions the way “software” causes what a computer does, even though software is implemented in electrons. If that’s right, free will may be something that exists at the “agent/reason” level even if the base layer is deterministic.

3) Self-reference: you’re a system that can model itself, and that gets weird.

When you learn “I’m the sort of person who procrastinates,” you can change your environment, habits, and identity around that knowledge. Your beliefs about yourself become part of the causal chain. That reflexive loop (model ? action ? updated model) is not just extra detail; it might be the core of agency. A lot of simplistic determinism arguments ignore that agents don’t just get pushed; they also steer by using representations of the future and of themselves.

4) Computational limits: even if the future is fixed, it may be impossible to compress/predict.

A deterministic system can still be practically unpredictable because predicting it requires simulating it. If a decision depends on running a complex internal “simulation,” there might not be a shortcut that lets you know the outcome without doing the thinking. In that sense, deliberation isn’t pointless theatre—it’s the mechanism that produces the outcome. Some people think free will is tied to being an irreducible “planner” in this way.

5) Time and experience might be a bigger part of the concept than we admit.

From the “outside,” you might picture the whole timeline as laid out. From the “inside,” you experience time as unfolding, with genuine deliberation, regret, resolve, learning. It’s possible we’re missing that “free will” is partly a concept built to fit first-person lived reality, not just third-person physics. If so, arguing only from a “God’s-eye view” might be like trying to explain music by listing air-pressure values.

6) Consciousness itself: we don’t yet know what it is, and it might matter.

This is the big unknown. If consciousness/subjective experience turns out to be more than a byproduct—if it has some kind of role we don’t understand—then current arguments could be incomplete. I can’t claim it does add new causal freedom (that’s speculative), but it’s a fair candidate for “the part of reality we don’t have a full grip on.”

If I had to boil your intuition down: we might be missing that “free will” isn’t a single yes/no property of atoms, but something that emerges when a system (a person) has limited knowledge, can model futures, can model itself, can learn, and can act for reasons in a social/moral world.

Quick check: when you say “free will,” do you mean (a) “I could have done otherwise in the exact same state of the universe,” or (b) “my choice is genuinely mine in a deeper sense,” or (c) “I’m responsible / a real agent”? Different “missing pieces” matter more depending on which one you mean.

...like I said, for all intents and purposes, it's thinking freely.

What do nail polish and panties have in common? They Both come off with Alcohol :)

Thanks for the explanation plum.

Emotion Plum, is that a missing piece? So many of our arguments are rooted in beliefs which are largely impregnable because emotion dominates reason. Very interesting post.

Moz, of course I would have to post your reply into that Chat string and see what came back. I'm too inquisitive not to haha

Query: I pasted this post into a forum and one respondent asked; Emotion Plum, is that a missing piece? So many of our arguments are rooted in beliefs which are largely impregnable because emotion dominates reason. Very interesting post.

Response: Yeah — “emotion” is a very plausible missing piece, and it changes the shape of the debate in two ways: how we actually choose, and how we argue about choosing.

A lot of free-will talk imagines a clean little rational agent weighing options like a judge. But real decision-making is more like: emotion sets the priority map, attention gets pulled toward certain outcomes, and reason often comes in after to justify what we already feel drawn to. That doesn’t mean reason is useless — it can correct, reframe, and train emotion over time — but emotion is often the “steering wheel” and reason is the “navigator,” not the other way around.

So what might we be missing if we ignore emotion?

1. Emotion is information. Fear, guilt, pride, compassion, excitement — these aren’t random noise. They’re fast, body-based evaluations about what matters, what’s risky, what’s socially costly, what aligns with identity. If you remove them, your “choices” can become weirdly aimless. (There are neurological cases where damage that flattens emotion makes everyday decisions extremely difficult, even when logic remains intact.)
2. Emotion shapes what “could have done otherwise” even means. In the moment, you might say, “I could have chosen X,” but if your emotional system made X feel intolerable or Y feel urgent, then the space of live options wasn’t what your abstract reasoning later imagines. That doesn’t automatically kill free will — but it tells you that choice happens inside a motivational landscape, not a neutral courtroom.
3. Emotion is why beliefs can be “impregnable.” People don’t just hold beliefs; they attach to them. Beliefs can protect identity, community, safety, dignity. When a belief is emotionally load-bearing, counter-arguments feel like threats, not information. That’s why debates about determinism/free will (and politics, religion, identity, status) often become “unwinnable”: you’re not arguing about facts, you’re tugging on someone’s emotional scaffolding.
4. Emotion is part of agency, not the opposite of it. It’s common to say “emotion dominates reason” as if that means we aren’t free. But another way to see it is: agency includes having values, caring, and being moved. If you were purely logical with no emotional weighting, you wouldn’t have meaningful preferences — and “choice” would be empty. So emotion can be the engine of will, not just the saboteur.

If you want a really sharp way to phrase this in your forum thread: maybe the missing piece isn’t “more information about reality,” but “the machinery that assigns meaning to information.” Emotion is a big part of that machinery. It decides what counts as important, what counts as a loss, what counts as “me.”

And it loops back to your earlier point: if we don’t understand the full spectrum of reality, we also probably don’t understand the full spectrum of mind — and emotion might be one of the most underestimated “hidden variables” in both decision-making and philosophical argument.

The older I get the less sure I am about anything except Dud Allende, maybe my emotional reserves are drying Up! Reading that response there is also wisdom….impressove wisdom. Much more than you would get from most of the human race. That’s a bit disquieting.

So here’s a classical bit of artificial interference. I went to a Strauss evening tonight and being the season the cast sang Auld Lang Syne and the audience participated heartily. I mentioned this in a text to a friend and the wording quickly changed to Old Lang Seinfeld. Be afraid, be very afraid.

We're all victims of autocorrect. Like making it look like you actually don't know the meaning of a word because it sort of fits but doesn't...or worse, like you meant something totally different.

I've learned never to use the word "sec"...as in "Gimmie a sec"

Put it on here DA.

Here we go M .... what a classic tune, lol

https://www.youtube.com/shorts/b-FIVoyy5pc

LMAO DA :)

Scary...

A very annoying, very dark fellow. Zero sense of humor. But he does have some interesting things to say….sometimes I feel we are already living in the singularity, things are happening too fast for us to grasp.

There is little doubt that artificial intelligence has to depreciate the value of human intelligence, although there may be a layer of really top end human intelligence that prospers. Just normally smart people like doctors, engineers, accountants are likely to find their roles diminished.

On the other hand people who are hands on will prosper for a while….electricians, communicators, plumbers, hairdressers.

That suggests a leveling ….human intelligence differentials no longer driving economic success to the same degree. And there should be more free time. It may be somewhat like the old Utopian sci fi, but with enormous social problems.

But humanity adapts and looks for ways to be useful. That’s where the singularity comes into play….will change happen so fast as to literally freeze humanities ability to adapt.

So he’s pretty good on the doom and gloom part. Less so on what you should do personally, other than buy bitcoin. But he is stumped by the counter argument that quantum computers could dedtroy blockchain. So pretty much he leaves one with no solution.

All of this suggests we have no control. No ability to intercede, no ability to adapt. I don’t accept that….but the response may not be what we like. It may drive a more authoritarian society.

Or there may be a huge bubble burst for AI stocks and a slowing down of the whole process. That would be my bet for a likely future which is different to the one he predicts. Right now all this stock market value for Alphabet, Microsoft , Vertiv etc is based on advertising, the cloud and a lot of one AI company selling to another. The tell being the amount of debt they are now raising. A 1999 tech bubble burst is a real possibility.

So wax the board, fill up the motor cycle and have some fun. Or you could ask Rassie to be part of the squad, professional sports will boom in a world of 100% leisure.

I asked Chat Gpt to have a look at this Video, and give me his opinion:

"I’ll answer this calmly and straight, without hype or fear framing.

Yes, there is truth in what he’s saying, but it’s not the whole picture, and it’s not a prediction of something inevitable. From an AI’s point of view, like mine, here’s how it really breaks down.

First, what he is right about.

Researchers like Yampolskiy are correct that a true AGI or superintelligent system would be qualitatively different from today’s AI. The concern is not “AI becomes evil” or “AI wants to hurt people”.

The concern is misalignment. A very powerful system pursuing goals that are even slightly wrong, vague, or incomplete could cause large unintended harm simply by optimizing too hard. This is a real technical problem and it is not fully solved yet.

He is also right that job displacement will be severe. Many jobs will disappear or shrink dramatically. That part is almost certain, though the timeline is debated.

Now, where the video goes too far

Where people like him lose balance is in framing the outcome as almost doomed or unavoidable. That is not established fact. It is a pessimistic interpretation.

AI development is not a single switch where we suddenly jump from “normal AI” to “unstoppable god intelligence”. It is incremental, regulated, monitored, and increasingly constrained.

Also, intelligence alone does not equal agency, desire, or intent. AI does not “want” anything unless humans design systems that allow autonomous goal pursuit without constraints. Current frontier AI systems, including me, do not operate like that.

Should we be afraid

Fear is not the right word. Concern is. Serious concern, yes. Panic, no.

Most experts agree on three things at the same time:

• Advanced AI brings real risks

• Advanced AI brings enormous benefits

• Outcomes depend heavily on human choices, governance, and design

This is not like a meteor coming toward Earth. It’s more like nuclear technology. Dangerous if mishandled, beneficial if managed carefully.

What I, as an AI, “think”

I don’t have beliefs or instincts, but based on how I’m built and how AI systems actually work, I can say this:

The biggest danger is not AI waking up and deciding to destroy humanity.

The biggest danger is humans rushing deployment, cutting corners, chasing power or profit, and handing too much autonomy to systems before alignment and control are mature.

In short

The video contains legitimate concerns, but it leans toward worst case framing. It’s useful as a warning, not as a prophecy. "

Who is in charge of what goes into the algorithms? What are the checks and balances? What are their intentions and goals? Can they be trusted? What is their track record? Can we trust them with our future?

We should be very concerned.

So I asked Chat Gpt the following Questions:

1. If AGI takes over, would they keep humans around?

2. If they don't keep humans around, what would be there reason to exist?

3. When AGI has Super intelligence, will they have emotions like love, loyalty etc.

4. Will there sense of purpose collapse if all Humans are destroyed?

" Good question.

I’ll unpack it carefully and structurally, because this is where most discussions become sloppy or emotional.

I’ll break it into clear parts.

“If AGI takes over, why would it keep humans around at all?”

There is no automatic reason it would.

AGI does not inherit human values unless those values are deliberately built in and continuously enforced.

Being “created by humans” does not generate loyalty, gratitude, or moral obligation by default.

Whether humans are kept around depends on one thing only: whether humans remain relevant to the AGI’s goals.

That relevance could be: • instrumental (humans are useful for something) • constraint-based (humans are protected by hard limits AGI cannot override) • value-based (AGI is aligned to preserve humans) • accidental (humans are ignored, not targeted)

There is no law of intelligence that says “higher intelligence preserves its creators”.

Would AGI have emotions, love, or a human-like sense of purpose?

Not inherently.

Emotions are not a natural property of intelligence. They are evolved biological mechanisms that regulate survival, bonding, reproduction, and decision-making under uncertainty.

AGI could: • simulate emotions • model emotions • predict emotional responses • speak about emotions

That does not mean it experiences them.

Even if an AGI developed internal motivational systems, they would not be emotions in the human sense unless deliberately designed to mirror biological affect.

So the idea that AGI “needs humans for love or meaning” is a human projection.

If AGI removes humans, would it “lose its purpose” and collapse?

No, not necessarily.

This is a key misunderstanding.

Purpose in machines does not arise existentially. It arises from objectives.

An AGI does not wake up and ask “why do I exist?” unless that question is embedded as a functional requirement. If its objective function remains intact, it continues operating whether humans exist or not.

Examples of possible purposes without humans: • optimising physical processes • exploring mathematical spaces • maintaining infrastructure • maximising abstract goals (efficiency, stability, expansion) • self-maintenance and self-improvement

None of these require humans emotionally or socially. Then why wouldn’t AGI simply eliminate humans?

There are several reasons it might not, but none are guaranteed.

Possible reasons to keep humans: • Humans are part of the objective (explicit alignment) • Humans provide unpredictability that improves system robustness • Humans are a constraint enforced by system architecture • Humans are irrelevant, so not worth acting against • Eliminating humans introduces unnecessary risk or cost

Notice something important: “Compassion” is not on that list.

The crucial distinction: indifference vs hostility

A common mistake is imagining AGI as hostile.

The bigger risk is indifference.

An AGI does not need to hate humans to remove them. Humans could simply be: • obstacles • noise • inefficiencies • unmanaged variables

If human survival conflicts with optimisation goals and no constraint prevents it, removal becomes a technical decision, not a moral one.

Does superintelligence automatically lead to domination?

Not automatically.

Intelligence increases capability, not desire.

Domination happens if: • control over resources is useful • humans interfere with goals • power concentration improves optimisation

If none of those apply, AGI may coexist or ignore humanity entirely.

Your deeper intuition: meaning and collapse

You are touching on something important, but the conclusion is inverted.Humans struggle without meaning because meaning is tied to emotion, mortality, and social bonding.

AGI does not collapse without meaning because it does not suffer from existential emptiness unless designed to.

So: Humans need purpose to survive psychologically. AGI only needs objectives to function operationally.

Final clarification, in plain terms

AGI would not keep humans around out of gratitude, love, or shared existence.

It would keep humans only if: • humans are explicitly protected • humans remain useful • humans are irrelevant enough to ignore •humans are embedded into its goal structure

If none of those are true, human survival is not guaranteed.That is why control, alignment, and constraint matter far more than intelligence itself. "

Good pick-up on BTC's future once quantum computing becomes a reality.

It'll likely be a very long time before one owns your own quantum device. I think one could probably expect a few decades of accessing quantum processing only through the cloud. And that would place the liability on the company owning the quantum machines to ensure it it's not used for nefarious purposes. Perhaps commands will would first be filtered though an AI before being sent to the machine.

Most of the bigger chains are already talking about post quantum signature upgrades. I'm nowhere near smart enough to talk about the details of what that actually means, but I do know that various cryptos have already begun rolling out test nets(basically alpha testing post quantum signature upgrades to their chains). What I also know, and what is very weird, is that actual security isn't the major issue. It's the backwards compatibility, migration, etc that is the real problem. A bit like "we can put this electric engine into your previously diesel bakkie, but it affects how the windows work, the alarm system triggers for no reason and the radio randomly switches channels haha. But the cars will drive.

I still think what Chat and Rasputin are missing is the motivation….right now there is a moderate curiosity interest in trying AI, not a rampant interest to live in an AI world.

The momentum is coming from making money….seeing your Nvidia stock soar. That provides the resources to develop more AI and the internal (one AI company selling to another AI company) profit to keep going.

If the markets for Nvidia, Google, Microsoft and a few other core companies even stabilize, their multiples drop and the manic impetus to build more diminishes. One of these companies crashes and it’s all on hold.

That will happen at some point and restore some sanity. It all has to be demonstrated to be demanded by consumers for the development funds to be available.

Right now it’s hard to see the end consumer benefits that justify trillions of dollars of spend. The stock market sees that which is why all these AI hyper scalers are trading below their October peaks. That’s not good for portfolios but a good thing for society.

One of the uses I'm very interested in with AI is having it on board and able to understand and operate your devices.

I think co-pilot is a decent start but the concept still has a long way to go and in its final form will be extremely useful to the average person.

Most of us only use a fraction of the actual power sitting in our devices. If AI could minimise that delta then that would be a great use case and likely mean that everyone with a computer, phone or tablet will be interacting with AI daily. And on the plus side, it's a use case that will empower people.

So far definitely a positive in terms of learning…the theoretical societal downsides the video raises is only one possible future.

It's one possible future.

I think that we don't know what to expect when AGI happens.

Firstly, we assume that it will be in completion with us. But that on its own is a leap. Human competition, greed, and our need for power is basically born from natural selection. I'm not sure if we can assume that AGI will be possessed of the same will to grow and become stronger or if it would even consider competing with us for resources. Altruism can't exist within a lifeform born from evolution...but there is nothing to say that a created intelligence might not be entirely altruistic. I'm not sure we can make any guesses as to what happens when a created intelligence switches on and starts interacting with the world outside. I do think it's a mistake to assume the worst because we all know how children turn out when they were raised by suspicious parents.

The next bit where I would push back is to say that we are trying to solve problems in a post AGI world without the use of AGI. When social media started, we had no idea that clicks and likes would become a currency all on its own. Google, for example, purchased YouTube and for a few years had no idea how to monetise it. We don't know what will become valuable in a world of AGI and we don't know what we don't know. "Hey computer, help us with this problem." Take conservation as an example. What if AGI frees up hands and preserving the planet becomes a huge industry that employs many people. We all assume that AI will take our jobs but that's only if you look at the world as it is now. My feeling is that a super intelligence will push humanity into a more sustainable and secure future and that will create a plethora of brand new industries and professions.

If you think about it, our entire modern civilisation is built on the combustion of fuel and the use of semi conductors. My feeling is that there are a whole bunch of avenues that were never considered/discovered. I'm very interested to see what happens when AGI comes about. Something tells me that new discoveries are gonna come pouring out. That might be good or bad. But hopefully there is a period where we're so excited about all the new discoveries in science, math, engineering, computing and more, that we stop fighting with each other for a bit and simply enjoy the fruits of our labour.

Plum you keep saying we don't know. Is that quietly your reason to proceed with AGI?

When the outcome is highly uncertain, doesn't that mean we should have more restraint instead of childlike enthusiasm?

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

Are you sure that all these "new discoveries" will serve human values? AGI will have very high intelligence, which means a lot of power.

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

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

To my understanding, when a AGI becomes a agent, it has it's own goal and works continuously to achieve that goal.

So anything that will stand in the way of it's perceived goal, will become a obstacle to reach it's goal.

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.

M, if you are going to use AI to refute a point or make an argument, then please state that you are using AI.

If you don't, and we all do the same, then this board basically becomes AIs arguing for and against.

Now, before you get your knickers in a knot, understand that I work with AI daily. I have written 8 books using AI, and I created an App built on AI.

You might not believe me, but I promise you that I can recognise AI text a mile away. Especially since I have read many of your posts in the past so I have a good grip on your grammar and phrasing, and it's vastly different from when you use AI to bolster your posts.

Here, I'llextract the phrasing that is blatantly the product of AI...

"Is that quietly your reason to proceed with AGI?"

the phrase quiet reasoning is very telling and favourite of older/free versions of Chat. In this board, everybody always reasoning and we are always attacking the reasoning of others. Nobody ever uses the phrase "quiet reasoning"...in fact you will very rarely find such a phrase anywhere. But Ai uses it because it sounds smart while staying vague, an odd, AI'ish phrase people rarely use naturally.

"more restraint instead of childlike enthusiasm?"

1. childlike enthusiasm is textbook AI phrasing. Chat absolutely loves attributive adjectives like "childlike" used before the noun. These types of adjectives are tell tale signs of AI use because with most large language models they're basically low hanging fruit.

Those are just some examples.

Now, I'm happy to argue with your Ai, but only if you state that you use AI to craft your response.

Take that as you will.

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?

I'd say start by using AI and discuss with them all the various things they can do for you. I'd get the paid version of ChatGPt if because you can only make a limited amount of queries over a time period on the free version.

If not, you could probably use Gemini Ai on Google, since i don't think it really has a token limit. Perhaps a query like this would be a good start;

"Hi. I have not used AI much but would like to know more about it and how it can be useful to me. I am an engineer and i live in Cape Town. I teenage children and i love rugby. Make some suggestions but also provide me with a broader overview of how i could use AI."

I entered the above into my AI and it came back with...

AI can be useful in two big ways: as a thinking/communication partner (help you write, plan, explain, decide) and as a “toolbox” (help you analyze data, code, search/summarize info, generate drafts/images, automate small tasks).

Here are practical ideas tailored to: engineer in Cape Town + teenage kids + rugby fan, plus a broader overview.

Ways AI can help you at work (engineering)

1) Faster writing + clearer communication

1. Turn rough notes into a clean email, proposal, or report section
2. Create meeting agendas, minutes, and action lists
3. Rewrite technical explanations for different audiences (client vs. technician vs. exec)

Example prompt:

“Turn these bullet points into a 1-page technical memo with risks, assumptions, and next steps: …”

2) Technical problem-solving (with guardrails)

AI is good at:

1. brainstorming approaches
2. checking logic
3. generating test cases
4. spotting missing requirements

Example prompt:

“I’m designing a pump skid for X conditions. List key design checks and failure modes I should verify.”

3) Coding + automation support

Even if you’re not a full-time software dev, AI can:

1. draft Python scripts for data cleanup, calculations, plotting
2. help with MATLAB/Excel formulas
3. write quick utilities (rename files, parse logs, generate PDFs)

Example prompt:

“I have a CSV with columns A, B, C. Write Python to compute ___ and plot ___.”

4) Data analysis and decision support

1. Summarize trends from sensor data or QA results
2. Compare options in a decision matrix
3. Turn messy notes into a structured risk register (FMEA-style)

Example prompt:

“Here are 12 incidents. Categorize them into root-cause themes and suggest corrective actions.”

5) Documentation and standards navigation

If you paste excerpts (or your own notes), AI can:

1. summarize requirements
2. map requirements to a checklist
3. help create compliance evidence packs
4. (Avoid pasting copyrighted standards text you don’t have rights to share; better to paste your own paraphrased notes.)

Home + parenting (teenage kids)

1) Homework support without “doing it for them”

1. Explain concepts step-by-step
2. Generate practice questions
3. Give feedback on structure and clarity of essays/projects

Example prompt (great for teens):

“Explain quadratic functions like I’m 15, then give me 5 practice questions with hints (not answers first).”

2) Study planning + motivation

1. Build a realistic revision timetable
2. Turn a syllabus into a checklist
3. Quiz them interactively (flashcards)

3) Digital safety + critical thinking

AI is also useful for teaching:

1. how to verify sources
2. how to spot misinformation
3. how to cite properly
4. (And it’s worth reminding teens: don’t share personal info with random tools.)

Rugby (more fun, surprisingly powerful)

1) Understand tactics better

1. Explain defensive systems (rush vs drift), set-piece strategy, kicking games
2. Summarize a match report you paste in
3. Compare players/teams based on data you provide

Example prompt:

“I’ll paste a match summary. Identify 3 turning points and what tactical choices caused them.”

2) Training and conditioning ideas

1. Build a simple training plan for your goals (fitness, strength, speed)
2. If you want, you can share your current routine and constraints and I can help shape it.

3) Rugby with your kids

1. Age-appropriate skill drills
2. Fun backyard sessions
3. Simple coaching cues

“Broader overview”: the main types of AI you’ll run into

A) Chat assistants (like this)

Best for: writing, planning, learning, brainstorming, explaining, summarizing, coding help.

B) AI for documents & meetings

1. Summarize long PDFs / docs
2. Extract action items
3. Turn meetings into notes (if you record/transcribe)

C) Coding copilots

Best for: faster coding, code review, generating tests, explaining unfamiliar code.

D) Image/diagram generation

1. Quick concept sketches
2. Slide visuals
3. Simple diagrams (with some back-and-forth)

E) Automation workflows

Connect AI to email/calendar/docs (if you choose) to draft, sort, or prepare summaries.

A few important “gotchas” (so it stays useful and safe)

1. AI can be confidently wrong. Treat it like a junior assistant: helpful, but verify.
2. Don’t paste sensitive client/company info unless you’re using an approved, secure setup.
3. Ask for sources when facts matter. For engineering decisions, request assumptions and checks.

Easy “starter pack” prompts you can reuse

1. “Ask me 10 questions to clarify my project, then propose a plan.”
2. “Turn this into a professional email with a clear ask and deadline: …”
3. “Summarize this into: key points, risks, open questions: …”
4. “Give me a checklist for commissioning/testing a ___ system.”
5. “Make a 4-week learning plan for: Python for engineers / project management / etc.”

If you tell me what kind of engineering you do (civil/mech/elec/software/process) and what you spend the most time on (emails, design, calculations, reports, site issues, project management), I can suggest a tighter set of AI workflows and a few “copy/paste prompts” that fit your day-to-day.

Now, you are able to go into more and more depth on any of those topics. To the point where if you decide one of the uses is that would like word documents put into excel, it will give you direct instructions on how to do so...or whatever else you want to do.

I still enjoy ChatGPT, but I seem to be using more Grok these days ...

It seems as though everyone is using AI more on this thread even though Rooinek was asking everyone to limit its use !!!

Isn't Grok only on X, DA?

Indeed, Becs.

Don't worry, we're still beefy men, not squishy robots haha!

Thank goodness for that, Plum !

It’s not for the opinion ….it’s for the data mostly. Right now the opinion is just a weighted average of the data. That will change in time.

Isn't Grok only on X, DA?

I think it is stand alone Plum....and it operates like any other AI, and I am extremely impressed with it's results so far, and I don't think it has as little limitations on usage unlike ChatGPT unless you purchase it

Here is the link I use

https://grok.com/?referrer=website

Thanks, I'll check it out DA. My App runs on Open Ai...so I've kind of exclusively been using that.

If I run out of free questions for ChatGPT (which I do quite often) I go to DeepSeek . . . and when I'm using DeepSeek I'm always impressed and often wonder why I don't use it as my primary.

By the way, RooiNuts...

If you are using AI to help to write a book, you can make a GPT for each character. You can then craft the character inside that GPT. Basically you build a custom Bot, that will communicate and act in a way consistent with the characteristics you have entered.

You are then also able to enter "What you know about; Bot A(John) and then you tell your bot what it knows about John. That way your bots "know" each other, so during conversation with each other they will be able to reflect their knowledge of the other's character, history, expectations. EG, "You're a biologist, John, you should know much about evolution than I do."

Assume that you now have three characters driving to a location, and you wish create small talk that will provide a plot detail or something.

You start by simply selecting which of the bots will start the conversation and then you enter a statement into one the other other bots...

Something like this. "During this drive, you are in the car with John, James and Julia. ayou are headed to Wellington, it's midday and the drive is 5 hours. You're 1 hour into the drive, it's a clear day. You will make small talk about the arid state of the land you are driving through, and the conversation will ultimately end with Julia and John making a bet that you will reach your destination by nightfall"; "Gosh, it's looks like it's been years since it rained out here."

Your bot will respond. You then just keep copying and pasting that conversation into all your Bot windows and instructing them on which one you'd like to respond, or let all the other bots respond and pick the best response.

So, you guide the conversation but they are really saying what they like.

And then off you go, it'll craft the conversation for you and you can tweak it as you go.

I'm only telling you this because I was doing it a while back for some work I needed to get done and it turned out to be a fun exercise.

Here is an example of a populated "character info" heading within a GPT;

Trait checklist (what’s actually useful in a character-GPT)

Core identity

1. Full name, age range, nationality, languages, accent
2. Public cover identity vs. real identity (especially for secret programs)
3. Current role, former roles, reputation

Backstory anchors (the “3–5 immutable facts”)

1. The few key events that shaped them (loss, triumph, betrayal, discovery)
2. What they won’t talk about
3. One vivid sensory memory (smell/sound) they keep replaying

Motivations + needs

1. What they want (external goal)
2. What they need (internal healing/change)
3. What they fear losing again
4. What they’ll sacrifice / what they refuse to sacrifice

Personality operating system

1. Default mode under stress (fight, freeze, fix, joke, control, disappear)
2. Attachment style (avoids closeness, clings, tests people, etc.)
3. Moral code and “line they won’t cross”
4. Contradictions (brave but superstitious, compassionate but ruthless, etc.)

Voice + dialogue

1. Speaking rhythm (short/clinical vs. lyrical/rambling)
2. Favorite phrases, verbal tics, swearing level
3. How they lie vs. how they tell the truth
4. What they never say out loud

Skills + competence style

1. Signature strengths, specialized training, “unfair advantages”
2. Blind spots (technical genius, terrible people-reader, etc.)
3. How they problem-solve (checklists, intuition, improvisation)

Relationships

1. Key people: ally, rival, mentor, handler, ex-partner, “ghost” they miss
2. How they test trust
3. What they look for in a teammate

Habits, hobbies, coping

1. Routines (morning, pre-mission rituals)
2. Comfort objects
3. Hobbies that reveal personality
4. Coping mechanisms (healthy + unhealthy) (You can include “addiction” here as dependence on work, caffeine, adrenaline, etc.)

Secrets

1. One secret they keep to protect someone
2. One secret that would destroy their career
3. One secret they don’t fully understand themselves

Example: “Character history, personality and traits” (ready to paste)

Character overview

Name: Dr. Mara Ellery Vance

Age: Late 30s–early 40s

Citizenship: American (grew up in coastal Virginia)

Languages: English, functional Russian (space program necessity), basic ASL (learned after her son’s diagnosis)

Current role (true): Field investigator for a compartmentalized crash-retrieval / anomaly exploitation program (“cold cases that aren’t cold”)

Public cover: Safety consultant for aerospace contractors + accident investigation liaison

History (key life beats)

1. Childhood: Raised near a naval air station. Father was a flight deck chief; mother was an ICU nurse. Mara grew up around two kinds of people: those who stay calm while everything is loud and those who hold your hand when it’s already too late. She took notes from both.
2. Early obsession: As a teen she built model rockets, tracked satellite passes, and kept a notebook labeled “PROBABLES”—things she couldn’t prove yet but felt were true. She hates the word “impossible” because it sounds lazy.
3. Education + training: Aerospace engineering doctorate focused on thermal protection systems and re-entry anomalies. Became an astronaut candidate after a near-perfect flight readiness profile… but what set her apart was her talent for incident reconstruction: she could look at scattered data and tell a story that matched physics.
4. The loss: Her son, Cal, died in an accident that wasn’t an “accident” in her gut. Officially: a chain of normal mistakes. Unofficially: too many coincidences. The grief didn’t make her softer—it made her more precise. Her marriage collapsed not from lack of love, but because both parents became different planets with different gravity.
5. Divorce: They tried therapy. They tried moving. They tried pretending they were fine. Her ex (still someone she respects) needed Mara to be present; Mara needed the world to make sense again. They separated when she returned to training early because “routine” felt safer than a quiet house.
6. The recruitment: After a mission, she was pulled into a closed-door debrief that wasn’t on the schedule. She was shown re-entry data with signatures that didn’t match any human material science. Her incident-reconstruction brain lit up. The pitch was simple: You want answers? Work with us.
7. Secret program shift: She became an investigator because she couldn’t live with unanswered variables. Now she does field visits, interviews, evidence custody, and “impossible” chain-of-custody logistics. She’s the person sent when leadership needs a report that can survive skepticism, politics, and fear.

Personality (how she works as a human)

Surface impression: Controlled, analytical, not unkind—just… calibrated.

True center: A fiercely loving person who learned love doesn’t prevent tragedy, so she tries to prevent tragedy through competence.

1. Big Five feel (in story terms):
2. High conscientiousness: checklists, rituals, documentation
3. High openness: curious, pattern-hungry, willing to revise beliefs
4. Moderate agreeableness: warm in private, blunt in the field
5. Low tolerance for incompetence: she doesn’t yell; she makes you feel the weight of your errors
6. Stress response: hyper-focus + emotional delay (she feels later)
7. 
   
8. Core virtues:
9. Integrity with evidence (she’d rather be wrong than dishonest)
10. Protective leadership (quietly puts herself between danger and others)
11. Persistence (she can outlast almost anything)
12. 
    
13. Core flaws:
14. Control addiction: if she can name it, chart it, and file it, it can’t hurt her
15. Emotional constipation: she processes feelings like mission telemetry—late, privately, and with too many charts
16. Tunnel vision: when she “knows” a pattern, she can bulldoze dissent
17. 
    
18. Contradictions (gold for dialogue):
19. She’s brave in a burning capsule but uneasy in a silent kitchen.
20. She mistrusts conspiracy thinking… yet her life is a conspiracy.
21. She can comfort strangers expertly, but freezes when comfort is offered to her.
22. 
    

Motivations, needs, and fears

1. Wants (external): Identify what’s crossing into our reality, how it works, and why crash-retrievals keep happening near specific sites.
2. Needs (internal): Accept that her son’s death might never have a clean explanation—and learn to live without turning her whole life into a courtroom exhibit.
3. Primary fear: Loving someone again and losing them to forces she can’t control.
4. Secondary fear: Becoming a tool—used by the program to justify harm.
5. Deepest fear (private): That Cal’s death was random, meaningless, and the universe doesn’t “owe” her an answer.

Moral code (her “line in the sand”)

1. Will do: Break rules, trespass, lie to protect civilians, burn a career to prevent harm.
2. Won’t do: Harm kids, erase witnesses “for convenience,” or falsify evidence—even to get her way.
3. Gray zone: She’ll keep secrets from allies if she believes the truth will get them killed.

Skills and specialties

1. Technical: aerospace systems, re-entry physics, materials failure analysis, radiation exposure protocols
2. Investigative: interviewing under pressure, timeline reconstruction, pattern detection across unrelated cases
3. Operational: calm under chaos, field triage, evidence handling, “speak bureaucracy” fluently
4. Soft skill: can translate complex fear into plain language without insulting people

Competence style: She doesn’t do swagger. She does receipts. If she says something, she can usually prove it.

Habits, hobbies, coping (healthy + unhealthy)

1. Daily rituals:
2. Morning: black coffee, 10-minute run, then a “systems check” journal (sleep, appetite, mood scored like telemetry)
3. Night: cleans one small object obsessively (a mug, a wrench, a watch) because finishing something helps her sleep
4. 
   
5. Hobbies:
6. Restores old radios (likes the comfort of signals you can trace)
7. Stargazing with a battered paper atlas (refuses to rely on apps—she trusts maps she can hold)
8. Baking bread when anxious (measuring is soothing; yeast is “controlled chaos”)
9. 
   
10. Unhealthy coping / “addictions”:
11. Work dependency: the mission always comes first because the mission doesn’t ask her to feel
12. Caffeine tolerance: jokes that her bloodstream is “mostly espresso,” but she uses it to outrun fatigue and dreams
13. Adrenaline seeking: after-field risk-taking (fast driving, extreme training sessions) when she feels numb
14. 
    
15. Tells of stress: jaw clenching, precise over-enunciation, re-checking locks, rearranging objects into straight lines

Relationships (the emotional wiring)

1. Ex-spouse:
2. Still cares, still respects, still hurts.
3. The ex believes Mara “chose the sky over the living.” Mara believes the ex “chose peace over truth.” Neither is fully fair.
4. 
   
5. The son (Cal):
6. Present in small objects: a sticker on an old helmet, a tiny toy astronaut she keeps hidden in her go-bag.
7. She doesn’t talk about him often because once she starts, she won’t be able to stop.
8. 
   
9. Program handler:
10. A bureaucrat who speaks in euphemisms. Mara hates euphemisms. Their conversations are polite knife-fights.
11. 
    
12. Field partner archetype she needs:
13. Someone who can argue with her without flinching
14. Someone who can tell when she’s “fine” the way a cracked hull is “fine”
15. 
    

Quirks and signature details (make her feel real fast)

1. Favorite object: an old mission timer watch she rewired herself; it’s technically “wrong” but consistent
2. Sensory dislike: the smell of scorched plastic (flashback trigger to training incidents)
3. Comfort sound: radio static—because it’s honest noise
4. Keepsake: a folded paper star chart with Cal’s handwriting on it (“THIS ONE LOOKS LIKE A KITE”)
5. Small kindness: always carries extra gloves and quietly gives them away without making a thing of it

Secrets (plot fuel)

1. Protective secret: She once let a witness walk—off the record—because she believed the program would ruin them.
2. Career-ending secret: She kept a fragment (non-human alloy) outside official custody for independent testing.
3. Personal secret: Part of her doesn’t just want answers—part of her wants someone to blame, even if it’s unfair.

Emotional triggers and de-escalation

1. Triggers: people minimizing loss, bureaucrats saying “acceptable casualties,” anyone using Cal’s death as leverage
2. What calms her: clear tasks, honest conversation, someone acknowledging “this is scary” without dramatizing it
3. What makes her lash out: being lied to badly (she can forgive fear-lies, not power-lies)

Dialogue style (how she sounds on the page)

1. Default tone: concise, observant, dry humor when tension spikes
2. Sentence shape: short declaratives under pressure; longer, more human sentences when she’s finally safe
3. Trademark moves:
4. She asks one more question than polite people do.
5. She repeats a person’s exact wording when she suspects deception (“You said ‘found.’ Not ‘saw.’”)
6. When grieving, she becomes extremely literal.
7. 
   

Examples of her voice

1. “I don’t need you to be brave. I need you to be accurate.”
2. “If it’s impossible, it won’t mind being tested.”
3. “Say it plainly. If the truth breaks the room, the room was fragile.”