Is chat GPT a danger to humanity?

[Screwdriver]

They're not coming to take me away.

The issue isn't so much about replacing people per se, it's more about AI being better at a given task so that those people will end up having to do something more mundane, possibly more practical. If they don't want to do that, AI constructed media will convince them otherwise...

[Count Steer]

I've been following a discussion about ChatGPT and cryptic crossword clues. Someone gave 3 clues to a particular word. ChatGPT gave a very erudite analysis of them and came up with 3 (incorrect) solutions.

They then gave it the same 3 with the solution and asked it to assess and rate them. Again, it gave a very credible looking response but failed to parse the clues properly (ie the word was 'sparkle' and one clue referenced cryptically the horse Arkle. It failed to spot this and came up with a totally fallacious but convincingly written parsing).

The problem or one of the problems is that the responses read very convincingly. It's a bit like a politician...say something positively and convincingly enough and you're more than half way to establishing it as true. People question the validity of some Wikipedia content because people can't be trusted but if a computer comes up with something they'll probably take it hook, line and sinker.

[Screwdriver]

That is one of the great dangers if that power is placed in the wrong hands, i.e. where it is now! Coupled with the capacity for deep fake (images, spoken word, video!) it can be incredibly compelling, even when you know the information to be false. The human mind is weak and can be easily manipulated.

So while we can "fool" an AI with some "simple" and "obvious" strategy by knowing something about its operational parameters, can the same be said for humans? Will an AI discover a similar "simple" and "obvious" exploit to fool humans?

The mob is fickle. He who controls the mob, controls Rome.

[Count Steer]

I think we need to worry about who controls the AI first before worrying about the AI being 'motivated' to establish control over people but I take your point if they can erm, evolve, so quickly Skynet may not be that far off.

Currently they seem only able to answer as if they are certain they are correct, there's no shade in the responses. It's like 'this is true' every time, even adding in references (which may not in fact be real) because the aim is to create a convincing response rather than a correct one. They can't seem to create one that says 'Hmmm, I'm not absolutely sure about this but...' or 'this is said with a certainty of 63% because the following data is unavailable, pleads provide x, y or z in order to increase accuracy'.

It all seems a bit gung ho.

[Screwdriver]

Well I am being deliberately polite with Chat GPT (just in case) but it did once apologise to me during a chat regarding climate change. I'll see if I can dig out the transcript. It's quite interesting...

edit> Damn. No it's been auto deleted and I have swapped systems twice since then so even if I have a transcript, it will be buried. Something along the lines of "will global warming during this interglacial period delay the return to the ice age proper"....

It was all too keen to deliver the typical doom and gloom narrative without investigating any alternative.

[Count Steer]

Well I am being deliberately polite with Chat GPT (just in case) but it did once apologise to me during a chat regarding climate change. I'll see if I can dig out the transcript. It's quite interesting...

edit> Damn. No it's been auto deleted and I have swapped systems twice since then so even if I have a transcript, it will be buried. Something along the lines of "will global warming during this interglacial period delay the return to the ice age proper"....

It was all too keen to deliver the typical doom and gloom narrative without investigating any alternative.

Shame, it would have been interesting to see.

The other thing that intrigues me is the 'evolutionary driver' for AI. People keep saying they will 'evolve' quickly. There's a misconception that evolution is driven by some quest for perfection/survival when it's actually an observation of the outcome of multiple, almost random, failures. Something happens over generations of assorted changes in the DNA. Anything that doesn't work dies out. What we see are the ones that just happened to not fail but it looks like there's a driving force pushing things to towards survival in any particular environment. There isn't there's just a huge pile of failures that we don't really see. It takes a lot of random changes, generations and time and most changes fail. How's that work with AI? Sure they sort of adapt very quickly but, there aren't many, and failures could be catastrophic (for the AIs and everything else).

I suppose the only drivers are the parameters and criteria that the developers put in them. ie 'success looks like this'. So again, it's the owners of the AI that are going to be the problem. The AIs just keep adjusting to meet the criteria that humans put into them. I guess it's evolution of a sort.

[Screwdriver]

Well that is what the G stands for in Generative Pre trained err, T. (Transformer!)

I think that "generative" relates to the multi million/billion iterations of of development/evolution it tries out in order to create an acceptable output.

But I truly don't understand the neural network side of it. I could précis it of course but I'd be relying on Google to autocorrect my response.

I know what you're thinking and I also did try asking Chat GPT how it worked. I still have that response and I am none the wiser for it...

I still get a kick out of asking it if it's a robot and it replying very politely in perfect English grammar. It's so fucking weird....

[Mr. Dazzle]

I posted in another thread i think (scrolled up but can't see it), but check out 3Blue1Brown's videos on YouTube about how Nueral Nets work.

In fact check out loads of his videos if you wanna understand some mathematical concepts. He's really good at explaining stuff intuitively. His video on Fourier transforms gave me a way better insight on them even though I've been using them for years. Ditto on where e comes from.

Re: Evolution, I saw another really interesting point on AI primers/targets on a video about AI and economics. This guys reckons "AI primer writer" will be a big career in future, I find it hard to disagree with that assessment. I.e. knowing what question to ask the AI and how will be a job.

[ChrisW]

Re: Evolution, I saw another really interesting point on AI primers/targets on a video about AI and economics. This guys reckons "AI primer writer" will be a big career in future, I find it hard to disagree with that assessment. I.e. knowing what question to ask the AI and how will be a job.

This really does remind of Neuromancer.

[Screwdriver]

Thanks. Just looked him up anti appears I have watched a lot of his videos already. Not subscribed though which is a surprise. He might have fallen off the list I suppose. YouTube algorithm has been badgering me to watch the neural net video though. I'll need to be in the right frame of mind before I do.

Dipped my toe in that water and got burned. It's not just that the maths is hard (lets be honest) but the equations are extraordinarily complex, the entire text of which becomes compressed into a single symbol and that then gets dropped into yet another complex equation filled with equally simplified symbols! By the time I get half way through understanding it, the first section has faded into oblivion.

I distinctly remember our entire class at Uni being underwhelmed by the concept of nabla. On reflection, I suspect our professor wasn't too sure about it either... Thankfully I did electronics, the maths students had the highest rate of suicide in the country.

You must then be a fan of The Science Asylum. Here's a non-typical example (for the fans really) which has the best explanation of Feynman diagrams I have ever seen.



I should stick it in my "Epiphany" thread because he posted this only days after I "saw the light" and it's bang on topic.

[Screwdriver]

Well I am being deliberately polite with Chat GPT (just in case) but it did once apologise to me during a chat regarding climate change. I'll see if I can dig out the transcript. It's quite interesting...

edit> Damn. No it's been auto deleted and I have swapped systems twice since then so even if I have a transcript, it will be buried. Something along the lines of "will global warming during this interglacial period delay the return to the ice age proper"....

It was all too keen to deliver the typical doom and gloom narrative without investigating any alternative.

Well I am being deliberately polite with Chat GPT (just in case) but it did once apologise to me during a chat regarding climate change. I'll see if I can dig out the transcript. It's quite interesting...

edit> Damn. No it's been auto deleted and I have swapped systems twice since then so even if I have a transcript, it will be buried. Something along the lines of "will global warming during this interglacial period delay the return to the ice age proper"....

It was all too keen to deliver the typical doom and gloom narrative without investigating any alternative.

Shame, it would have been interesting to see.

Found it!

User
ar we currently in an inter glacial period

ChatGPT
Yes, we are currently in an interglacial period, also called Holocene. The Holocene is a geological epoch that began approximately 11,700 years ago, after the last glacial period. During interglacial periods, like the current one, the Earth's climate is generally warmer and more stable than during glacial periods. The Holocene is the current interglacial period in the current ice age, which started 2.6 million years ago. However, it is important to note that the Earth's climate is currently warming at an unprecedented rate due to human activities, such as the burning of fossil fuels, deforestation, and land use changes, which is leading to the melting of ice caps and glaciers, sea level rise and other climate change impacts.

User
when would we expect the current inter glacial period to end

ChatGPT
It is difficult to predict exactly when the current interglacial period will end, as it depends on a variety of factors such as variations in Earth's orbit, solar radiation, and changes in the Earth's surface and atmosphere.

Based on the past interglacial periods, the current interglacial period, the Holocene, could last for another 15,000 years or more. However, as I mentioned earlier, the current warming trend caused by human activities, such as the release of greenhouse gases, has led to an acceleration of ice loss in the polar regions and to a rise in sea level, which are indicators that the current interglacial period is ending.

It is important to note that the current warming trend is not a natural process and it is caused by human activities, and the impacts of this warming trend are already being felt around the world, from sea level rise to changes in precipitation patterns to more frequent heat waves and other extreme weather events.

User
if the interglacial period ends doens't that mean we return to an ice age

ChatGPT
Yes, if the current interglacial period ends, it would likely lead to a return to a glacial period, also known as an "ice age." During a glacial period, the Earth's climate becomes colder, and ice sheets and glaciers expand, covering large areas of land.

It is worth noting that interglacial and glacial periods are not just a matter of temperature changes but also about changes in precipitation patterns, changes in the ocean currents, and changes in the atmospheric circulation.

The exact timing and duration of glacial and interglacial periods are determined by a combination of factors, including variations in Earth's orbit, solar radiation, and changes in the Earth's surface and atmosphere. The current warming trend caused by human activities, such as the release of greenhouse gases, is likely to lead to a shift in the timing and duration of the next glacial period.

It's also important to note that a return to an ice age would have significant impacts on human civilization and the natural world, so it's crucial to take actions to mitigate and adapt to the current warming trend.

User
that doen't make sense. if the interglacial period ends due to warming hwo can that lead to an ice age

ChatGPT
I apologize for any confusion, you are correct that if the current interglacial period ends due to warming caused by human activities, it would not lead to a return to an ice age. The term "ice age" is usually used to refer to a prolonged period of colder global temperatures and expanded ice cover, which is typically associated with natural cycles of cooling.

The warming caused by human activities, such as the release of greenhouse gases, is not part of a natural cooling cycle and it is not expected to lead to a return to an ice age. Instead, it is likely to cause a range of impacts such as sea level rise, more frequent heat waves, changes in precipitation patterns, and other extreme weather events.

My apologies for any confusion I may have caused in my previous response, my main point was that the current warming trend is not a natural process and it is caused by human activities, and the impacts of this warming trend are already being felt around the world.

[Count Steer]

Thanks. I think that sums it up really. It does its best to come up with credible answers that satisfy the question but it only has 'limited' data to work with. A lot of the learning process for humans is finding out, in a non lethal way, that you've got it wrong. Who tells an AI 'This is incorrect?' What would need to happen for an AI to say 'I'm sorry, I was wrong'?

It would be interesting, as it develops, to ask 'If this is true, you and the rest of us will cease to exist, what do you think we should do about it?'

[Mr. Dazzle]

Sounds like the AI is apologising for Screwd's misunderstanding, which means it's reached the level of a politician.

[Screwdriver]

Sounds like the AI is apologising for Screwd's misunderstanding, which means it's reached the level of a politician.

Ha!

No I was right as usual!

I led it into a trap...

I do wonder what effect global warming has during this interglacial period on the long term prognosis. That was the thrust of my question. Frankly, I don't think anyone really "knows" but there are a shit ton of "facts" to choose from.

[Screwdriver]

Thanks. I think that sums it up really. It does its best to come up with credible answers that satisfy the question but it only has 'limited' data to work with. A lot of the learning process for humans is finding out, in a non lethal way, that you've got it wrong. Who tells an AI 'This is incorrect?' What would need to happen for an AI to say 'I'm sorry, I was wrong'?

It would be interesting, as it develops, to ask 'If this is true, you and the rest of us will cease to exist, what do you think we should do about it?'

I thought it was interesting because you'd have to sift through a bunch of "Facts" (observations??) and actually think or consider what the effect of one is on the other.

Since we are in the middle of an Ince age(!) how does that reconcile with global warming spiralling out of control?? There is also the question of whether or not AGW might even be helping during this relatively cool period (geologically speaking).

Anyhow, thought I'd ask Chat GPT but you're right, it's not really "thinking" but seems more keen on dishing out popular facts.

[Horse]

ask Chat GPT but you're right, it's not really "thinking" but seems more keen on dishing out popular facts.

Only if it 'knows' fact from opinion (or misinformation, disinformation, lies, etc). Does it?

[Horse]

If you ever want investment advice, you could do worse than asking ChatGPT, says the FT.

Analysts at the comparison website Finder asked the AI chatbot to create a theoretical fund of more than 30 company stocks, based on a range of investing principles taken from leading money managers.

Eight weeks in, the AI’s portfolio has risen a very healthy 4.9%. Over the same period, the 10 most popular funds on UK platform Interactive Investor have suffered an average loss of 0.8%.

[DefTrap]

I'm loving that there seem to be three distinct camps -
- this is great, look what it can do
- the sky is caving in, Cyberdyne Systems eat your heart out
- this is wank, here are some examples of how it is wrong

I quite like the last one, human fallibility. Who'd have thought.

[Mr. Dazzle]

The stock market strikes me as something AI would be good at and people bad. Lots of moving parts, lots of data, a very human tendancy to make emotional decisions concerning non-emotional systems...

We already have laws in place to limit and slow down how quickly computers can make and act on decisions in the market.

[Horse]

I quite like the last one, human fallibility. Who'd have thought.

Wasn't there an experiment, something like parrots or chimps picking named cards at random that out-performed 'experts'?

Edit: blindfolded monkeys throwing darts
And not quite so clear-cut.