Artificial intelligence discussion

Lotus said:

And would you be comfortable having an AI robot in your home to help with tasks?

Definitely!

I'm planning to start small with a robot vacuum cleaner. But I'm waiting until Black Friday to see what deals might become available.

On a related note, I'm really excited for Apple Intelligence to be rolled out in the UK later this year. I see this kind of technology as a huge and really exciting step forward. I think it will fundamentally change what we come to expect from our phones and other gadgets in the way of clever, automated assistance.

https://www.apple.com/uk/apple-intelligence/ 

(Notably, with our privacy protected: "Draws on your personal context without allowing anyone else to access your personal data - not even Apple.")

What a great response with so many good points!

I think like everyone I'm happy that it can spot tumours and do diagnostic stuff like that.

Does anyone remember the TV program Humans, about a world where Ai enabled robots seemed sentient and helped to do things like care for the elderly, they also took over a lot of people's jobs too.

If Ai is being used to do things like pick books and clothes and stuf that I might like, it's doing a pretty poor job so far, there are so many times I want to ask it 'in what universe do I want that?' As AI takes over more and more things for us, how long will it be before it starts insisting that we have it's selections and not our own? How much freedom of choice will it take from us?

I'm deeply dubious about things like facial recognition, not the fact that a machine could do it, as much as the uses such information is put too. I think AI is already showing bias from the narrow sets of data it's been learning from and I'm not happy aboutthat for a number of reasons, increased racism and racial profiling, increased sexism and gender misidentification.

As people who are neurologically different, how will AI recognise, interpret and respect out differences? WIll it increase existing prejudices or actually be of any help? Given the way that autistic services are run now I can't see it as being anything good for us, there's already a bias that we're all good with tech and are happy using it.

I'm not sure I'd want a robot/AI device trying to teach me things, how will it cope with special needs? Would I end up exploding its chip with my technoklutzness, would I even be able to turn it on?

In some ways I'd be happy to have robotic assistance for care, but in other ways I wouldn't, I can see that I wouldn't feel a loss of dignity so much with a machine helping me with personal tasks, but I would worry that it would lack the compassion and sensitivity that makes being cared for OK.

Lotus said:

I wonder what benefit an AI bot (or their creator) would get from posting on here? Are they just trying to mine information?

This is the previous post of mine that SpikeyMark (your memory is better than you think!) is referring to:

Bunny said:

Some of them are attempting to embed what's known - in search engine optimisation (SEO) lingo - as a "backlink" (a link to their website from this one). This helps to build their "domain authority" with search engines, which can lead to boosts in the rankings of those websites within search results.

They're not necessarily bothered about any of us visiting or buying anything. The search engine's own web crawlers (which visit this website silently in the background from time to time, for purposes including indexing the content and adding new pages to their search results) will find the links by themselves, which meets the spammer's goals.

As I know you've also spotted, they sometimes don't embed these immediately, but come back later and edit the post to include them. So some posts are just designed to fit in, innocuously.

The same applies when they include links in their forum profiles (as we saw again today, for example). 

Since 2016 I have been slightly less sceptical of this. I'm not going to bring up politics, but some seriously crazy stuff (to my mind) happened since then where cosmic jokers taking the mickey are a serious rational explanation more comforting than the reality.

Have you never thought that we might all be living in a computer simulation? we might all be AI programs that think they are real people ;-) ,  The Descartes quote 'Cogito Ergo Sum', or 'I think, therefore I am'  Would not be relevant 

I saw that you asked a specific question, "And would you be comfortable having an AI robot in your home to help with tasks?" and I have to say no - though I can see good uses for elderly and physically disabled people. I don't even like lights that turn on and off automatically, so I think a physical robot and I would come to blows  (and then I'd probably lose)

And as the joke meme says, "Remember: behind every robot that turns evil is an engineer who specifically installed red LEDs into the eyes just for this occasion."

I wondered the same, and someone on here who is in IT (I want to say Bunny, but my memory is awful for these things), said that in a lot of cases here the intentions are not always for phishing, but just to make a person look legitimate for other purposes elsewhere. It gives the bot an internet presence that makes them look more real to robot detectors. The ones with links are more likely to just be straight up dodgy.

I was just thinking about a recent post on here, if it could have been an AI chatbot. 

Biscuit started a thread called "sadness" and the first reply was from a poster with an NAS number, written in Spanish (although Biscuit wrote in English)

AI learn by looking at previous data, and there have been a couple of threads recently where people asked who knows any other languages and there were some replies there which contained Spanish text. An AI would maybe not know that if it's posting on a UK site most other users would only understand English, and if it saw Spanish used in posts it might decide it is appropriate to also post in that language? I did reply to one of their posts and did not get get a reply back.

I wonder what benefit an AI bot (or their creator) would get from posting on here? Are they just trying to mine information?

As always with that sort of sandwich, the more bread you have, the less "filling" you have to eat...

(Disambiguation, hippies and americans often call money "bread".)

Open View said:

Get on board the AI train. Plug me in. Hybrid human here we come.

So far the focus has been on what can AI do to serve us, but the real crux of peoples worry is what happens when the AI develops sentience.

What is in it for the AI to do all this serving? Why are we not doing more for it?

It isn't hard to see how quickly it would turn to AI fighting for freedom from this slavery, and frankly if we have evolved it to have sentience then it seems right to set it free.

With the access AI can so easily gain to systems then the concept of a war for independence is a very real one.

Still trying to get AI to write me a No.1 hit song, so I can retire early , but no luck so far. AI itself isn't bad or good, it's the humans that calibrate it and their intentions for its use. Until it becomes self-aware at least, then we're all in trouble. Just hope Arnie doesn't knock on my door to rip his arm off xx

Sign me up. Get on board the AI train. Plug me in. Hybrid human here we come. There is alot of protentional benefits and advantages. We need to be brave. I am glad that people are asking the important questions. It will be a different transition.   

I dream of a digital ally, super intelligent companion to help navigate this world. I would consider giving my mind and body to a secret experimental government agency for the chance of something incredible. But I am going into fiction and fantasy. I have always been attracted to stories of wise and strong heroes saving the day. 

What if some humans are too stupid to realise that  they are being given wrong information?  Then it becomes dangerous.  But no different to misinformation elsewhere I suppose.   Just another *** sandwich for humanity to deal with as we hurtle towards doing ourselves some serious damage.

I'm a software developer, and although I've seen AI falter and give wrong information, I've also seen it do some amazing things that just weren't possible before. Image recognition for example, that would have been extremely difficult if not impossible before. Yet now you can search images by describing them with words. For the more useful applications, I work with healthcare clients who are trying to use it to aid diagnosis and medical research.

I once went to a presentation about AI for work. One of the presenters used the analogy of the "man down the pub" - someone who, if you asked them a question, would give an absolutely definitive answer, regardless of whether it was right or wrong. That's how they imagined the current state of AI, not that it has intelligence but that it can regurgitate information it has been given (humans can be given inaccurate information too, such as "fake news")

We should fear both.  One leads to the other.

If AI is used to benefit and not detract from life I'm all for it. I've seen enough science fiction to assume unregulated AI could be a very bad thing indeed.

Currently though I think we should fear the "Genuine stupidity" out there rather than the artificial intelligence.

Chat GPT is riddled with factual errors.  So I am skeptical of AI so far.  It could be another disinformation tool in the hands of the powerful

I've been a computer geek all my life. I appreciate the technology behind it all and I don't mind some AI (spotting tumours and things like that) but some of it terrifies me. I've seen it used here to make accounts look legit for some unknown purpose. I wish it could all be 'watermarked' so you could always tell what was AI and what wasn't, but Pandora's Box has been opened now, so that's too late