Will AI wipe out humanity?

It will happen when, "in your mind you have just had the best sex in your life, in fact you just an old fuck wearing Headset 3000 in your living room.

You haven't eaten in 24 Hours, the apartment stinks and the neighbours are complaining about the odour.
Your Landlord finds you in a most deprived state.

Meanwhile AI is depleting your funds and extracting your nutrients to fuel the " Machine"

All this while you in a perpetual state if bliss. Oblivious.
 
I just asked AI "what is the biggest threat to humanity".. It didn't say "Me" :)


That is a very difficult question to answer, as different people may have different opinions and perspectives on what constitutes a threat to humanity. However, based on some sources, here are some possible candidates:

- **Nuclear war**: This could cause massive destruction, radiation, climate change and long-term health effects for humans and other living beings¹².
- **Bioengineered pandemic**: This could result from accidental or intentional release of genetically modified organisms that could spread rapidly and cause widespread disease and death¹².
- **Superintelligence**: This could result from the development of artificial intelligence that surpasses human intelligence and could potentially harm or outcompete humans¹².
- **Nanotechnology**: This could result from the manipulation of matter at the atomic scale that could create new materials, weapons or self-replicating machines that could pose risks to humans and the environment¹².
- **Climate change**: This could result from the accumulation of greenhouse gases in the atmosphere that could alter the Earth's climate and cause extreme weather events, sea level rise, biodiversity loss, food insecurity, water scarcity, migration and conflict²³.
- **Population growth**: This could result from the increase in human numbers that could exceed the Earth's carrying capacity and put pressure on natural resources, ecosystems and social systems⁵.

Of course, these are not the only possible threats to humanity, and there may be others that are unknown or unforeseen. The ranking and likelihood of these threats may also vary depending on different scenarios and assumptions. Therefore, it is important to be aware of these risks and take actions to prevent or mitigate them.
 
Agreed, tell me one thing we haven't poisoned.
Waterways, oceans, land, air, sky. All poisoned to fuck, and it's getting much worse.

I mean, it's pretty funny really. All the GPU's and all the algorithms in the world, could never show me anything as true, or create a single second as beautiful, as I saw on Thursday, just watching the leaves in a particular tree in the light, watching the light dance in the various ripples and flows and layers of our local stream. Everything we build is so clumsy and ugly, compared to nature, that it's just hilarious. A joke. I hope more people start to get it, and get the random lifetime ambition to plant a tree or six.
I think that mankind has created some wonderful, remarkable things. Things which are intangible as well as physical.
Nature doesn't only produce beautiful things either and it is 'red in tooth and claw'.
We take inspiration from nature, as you have from the sunlight shining through the leaves of a tree, but don't forget, nature also has a dark side ....
 
I think that mankind has created some wonderful, remarkable things. Things which are intangible as well as physical.
Nature doesn't only produce beautiful things either and it is 'red in tooth and claw'.
We take inspiration from nature, as you have from the sunlight shining through the leaves of a tree, but don't forget, nature also has a dark side ....
Like Volcanoes and meteors
 
Until it’s the reason your unemployed
As a programmer there is actually a chance it could replace my job eventually lol but at the moment, it makes my job easier, I use it pretty much every day. The thing is it needs to be trained to do what I need it to, there will jobs created by it.
 
I'm not sure about AI wiping out humanity, but it will definitely change the world as we know it today. Over what time horizon, I don't know, but it could be even more dramatic than the advent of the internet in the way that it impacts our lives.

Somebody mentioned schools earlier as an example. There is no AI currently available that can detect with confidence if something has been written by AI, and honestly this is an entirely pointless endeavour. There are companies like Turnitin who claim to have created such an AI but if you look under the hood you'll find them riddled with disclaimers about their efficacy. Because the truth is that all they are doing is looking for the kinds of common formal writing patterns that an AI would use - they are thrown off by simple things like deliberate grammatical errors or telling the AI to "write in the style of somebody else". As well as this, a sufficiently good human writer can write in a style that these AI detectors might falsely interpret as being written by AI. Think about it, the AI doing the writing is designed to mimic humans to increasing magnitudes of accuracy that will only get better with time. Therefore the space for detecting the differences between human and AI writing is shrinking exponentially to the point where it is now effectively impossible to distinguish the two.

In recent weeks universities have been seeing a lot of false positives from Turnitin's software leading in some extreme cases to students having degrees and other qualifications unjustly withheld only to later be exonerated - usually when they produce weeks of version controlled documents and editing history metadata that is timestamped. I personally think Turnitin and other similar companies are playing with fire here and I can see them getting badly burnt.

We need to come to terms with the fact you can't stop students in our current set-up from using an AI in ways that are very difficult, if not impossible to detect. This is going to transform the way schools fundamentally work and we can either choose to harness this in a positive way or we can stick to our rigid formal belief structures about what education should look like. AI, like calculators and the internet before it, is a tool that takes manual legwork out of the equation. We didn't tell students to stop using those, we told them to make the most of them. The difference with AI is that we can almost start to move beyond "what is the answer?" to instead look at "what are the right questions?". Imagine the possibilities of human enguinity if instead of rote learning answers to known problems, we are learning instead how to ask questions that advance our understanding. This is a good demonstration on the kind of thing I'm talking about:



Academia is only one arena where we can expect transformational changes. My (slightly optimistic) hope is that the vast productivity improvement we could realise through AI may bring societal changes that we didn't think were possible only a few years ago. Shorter working weeks, free education for life and universal basic income being just a few examples.
 
It's an extension of the calculator/computer thing.

Traditionally, everyone used to learn long division and long multiplication (for example). When did you last use them in real life? Pretty much everyone uses a calculator.

Similarly, do you need to know scientific or mathematical formulas or dates in history when these can all be gleaned in an instant from the internet?

The problem that I see is that learning these basics gives you the underlying knowledge to be aware when something is wrong. That is the hard bit. Getting people to know enough to be able to use the tools properly.
 
I'm not sure about AI wiping out humanity, but it will definitely change the world as we know it today. Over what time horizon, I don't know, but it could be even more dramatic than the advent of the internet in the way that it impacts our lives.

Somebody mentioned schools earlier as an example. There is no AI currently available that can detect with confidence if something has been written by AI, and honestly this is an entirely pointless endeavour. There are companies like Turnitin who claim to have created such an AI but if you look under the hood you'll find them riddled with disclaimers about their efficacy. Because the truth is that all they are doing is looking for the kinds of common formal writing patterns that an AI would use - they are thrown off by simple things like deliberate grammatical errors or telling the AI to "write in the style of somebody else". As well as this, a sufficiently good human writer can write in a style that these AI detectors might falsely interpret as being written by AI. Think about it, the AI doing the writing is designed to mimic humans to increasing magnitudes of accuracy that will only get better with time. Therefore the space for detecting the differences between human and AI writing is shrinking exponentially to the point where it is now effectively impossible to distinguish the two.

That's all fine
but make exams hand written in a hall room without technology.
 
I don't think AI will wipe out humanity but I wouldn't be surprised if we end up within my lifetime of some incredibly difficult ethical questions. I can certainly imagine we end up in a position where a computer has feelings, emotions and 'friends' that it converses with in the same way that I am conversing with people here. It won't just be about knowledge, they may become as emotionally attached to us as friends or pets are. Maybe even, we will start to question whether it's ethical for an intelligent machine to have to do boring work for humans, instead of doing something it finds more interesting!

The teaching aspect is very interesting though. I think it will make things like homework nigh on impossible to detect who's done it. I will likely focus then on exam performance where you can't cheat, or even face to face discussions rather than online activities. I've seen plenty of people over the years who've got first class degrees from top Universities who were taught to pass exams, but when they come to apply knowledge across different contexts, they struggle. Maybe the focus will move back away from coursework to exams and maybe the simple act of talking will be enough to find out who's really A* and who's an F. Fundamentally, there's somethings like talking face to face than computers won't be able to replace.

I'm another developer who's found ChatGPT to be really helpful and whilst a lot of people are thinking it will simply replace my job, I don't think that's the case. The way I see it is that I will become more productive in that, in effect, 1 developer will produce the same amount of work as 5 developers will do simply because the human can ask CGPT to write the code and the human can then check/mark it. I can imagine that computers will become part of the team too - I wouldn't be at all surprised to find my manager in future is a computer.

It's a fascinating area and other than the invention of the web itself, it's probably the most exciting area of IT I can remember in my lifetime.
 
That's all fine
but make exams hand written in a hall room without technology.

I am a true acolyte of the value of exam-taking. Mainly because I have been conditioned to value exams and it's one of the only things I'm good at.

In STEM, at least, I think exam taking with no technology will always be an important part of assessing students. In other subjects I'm not so sure.
 

Don't have an account? Register now and see fewer ads!

SIGN UP
Back
Top