Nigella's Cream Pie
Fingerlickin good
Who needs real women?
Got something to say or just want fewer pesky ads? Join us... 😊
[Misc] Are you pro or anti AI?
- Thread starter Hovegull
- Start date
beorhthelm
A. Virgo, Football Genius
- Jul 21, 2003
- 36,065
it's not flippant, it's seeing the gap between expectation and delivery. GPT-4 model are 18mths old, GPT-3 which was a real game changer released early 2020. plenty of time to package up this tech into products, and sure there's some chat bots and industrial applications. so far the revenue from these is tiny % of the cost of running the models, running in tens of billions. thats the real elephant in the room, it doesn't look scalable, requiring more and more energy and GPU farms. the company executive say one thing, the balance sheets say something else. only one making money is Nvidia and the datacenter constructor (and hopefully nuclear plant builders, investing there).
- Oct 12, 2022
- 2,798
My prediction is twofold:
1. Increased specialisation. The robotics industry has already learnt this. It started out with tech demos and has evolved to providing “use-case specific” products.
2. Efficiency gains. Once the product is good enough to sell, work will take place to reduce power consumption requirements. If your AI agent has an energy cost comparable to a human employee, there’s not enough upside to completely refactor your business.
GPT-3/4 is largely tech demo. Specialised models already exist, and will be further refined until they are genuinely useful.
I agree that AI companies have not yet fully understood how to generate revenue, but they’re full of smart people. It won’t take long.
No thanks.
UnhingedSeagull94
Have a nice day….BANGBANG
- Jan 6, 2024
- 94
The issue with no jobs is that humans will be moved onto some form of social credit system. The govt. would control what you get to spend your social credit on. Social credit benefits would be decreased if you are deemed to be a bad citizen.
I’m 30 now and I dread and live in fear at the potential state of the world in the next 20 years.
I am very pro, but understand the impact on people and believe we all need to work to ensure we make it positive for all of us, which will not happen organically. Its current version is nothing like the hype, very limited and difficult to use effectively, but the opportunity for humans is huge. I do recognize though I like it probably because it is good at doing tasks I hate (mundane paperwork) and not very good at analytical work. If this changes I reserve the right to change my vote.
brighton_dave
Well-known member
- Apr 13, 2016
- 500
As someone who started out in the late 90s using a fax machine, I've been massively behind change and enthusiastic. It's been a stepping stone into a digital Product Manager that I work as today.
And then AI arrives, I have zero enthusiasm for it, and can't bring myself to feel at all interested. Loads of excitement in our organisation, and my scrum product team are really passionate, and all over wanting to drive forward with it. If only that had a PM who shared the same enthusiasm.
Unsure why, age perhaps?!
And then AI arrives, I have zero enthusiasm for it, and can't bring myself to feel at all interested. Loads of excitement in our organisation, and my scrum product team are really passionate, and all over wanting to drive forward with it. If only that had a PM who shared the same enthusiasm.
Unsure why, age perhaps?!
BrightonCottager
Well-known member
I'm very worried about it for lots of reasons, including:
1, how will my daughter (& millions others) get an entry-level job when she leaves University?
2, how do I tell if my students are cheating?
3, how can we stop it telling us what (some misogynistic, bigoted right wing nut job programmer thinks) we should do?
It's a scam to accelerate capitalism invented by a tiny minority of tech-bros. But most humans are too lazy to object.
1, how will my daughter (& millions others) get an entry-level job when she leaves University?
2, how do I tell if my students are cheating?
3, how can we stop it telling us what (some misogynistic, bigoted right wing nut job programmer thinks) we should do?
It's a scam to accelerate capitalism invented by a tiny minority of tech-bros. But most humans are too lazy to object.
Yes, it's a little bit later than 1984 but I think Orwell had the right idea.
A huge part of the issue with AI is that the term AI is currently being used as such a ridiculous catch all.
We’ve used AI for years for mobile banking, sorting our email, or simply logging into our devices. It’s just got considerably smarter all of a sudden.
As with all technological advances they are, in themselves, neither good nor bad, it’s how we use them that is the decider.
For me, so far, AI has remarkably improved my efficiency at work by assisting me in image editing and excel work. I’ve embraced a lot of things on my iPhone and think it’s going to improve my life hugely over the next few years. At the same time it’s going to have some bad uses, and leave plenty of people feeling left behind as technology continues to advance.
We’ve used AI for years for mobile banking, sorting our email, or simply logging into our devices. It’s just got considerably smarter all of a sudden.
As with all technological advances they are, in themselves, neither good nor bad, it’s how we use them that is the decider.
For me, so far, AI has remarkably improved my efficiency at work by assisting me in image editing and excel work. I’ve embraced a lot of things on my iPhone and think it’s going to improve my life hugely over the next few years. At the same time it’s going to have some bad uses, and leave plenty of people feeling left behind as technology continues to advance.
Wrong-Direction
Well-known member
- Mar 10, 2013
- 13,672
I'd love to see a robot crawl under your bath and replace a pair of 40 year old taps!
Razzoo
Well-known member
Teach
- Thread starter
- #53
And give shambolic service in a restaurant
Cotton Socks
Skint Supporter
- Feb 20, 2017
- 2,215
I'm not a fan. I'll just now ring a 'customer services' dept as chatbots annoy me & if the answer to my query was that simple I would've already found it out on the 'help' pages (which are also rubbish).
I tried the Southern Water chatbot the other day expecting it to be able to give the answer to a relatively simple question, nope I still had to ring them. Customer service chatbots have done a good thing for the local economy though in a very small way. I've had to buy 4 major appliances over the last 5 years, 1st was from Argos & trying to get out of the chatbot cycle to speak to an agent to say that my washing machine smells of burning was ridiculous. Next one Curry's & the dishwasher that they'd just installed was leaking, same thing. I browsed further than the 'big retailers' for my oven & spoke to a human at Carters they gave me sensible advice, fitted it perfectly & I had an 01273 number to call. I've just bought a gas hob from them, it's not been fitted yet as I don't want it yet. Again I was able to speak to someone about potential issues with sizing & when it's fitted I'm not going to worry that they don't know what they're doing. Curry's couldn't fit a dishwasher correctly I'm not going to trust them with gas! How is AI going to fit my cooker or dishwasher? How is it going to get it down my steps? How is it going to know the plumbing 'quirks' where I live?
I've also seen someone cheat their way through a degree with AI. They started using AI from the moment they started so progress in their essay standards have improved in line with their 'learning' from tutors & AI's learning. The degree topic is not something I would say was suitable for AI, I found it really bizarre why someone would want to cheat so much on something that is based on theories, rather than learn what the actual theories were & reach their own conclusion. To be fair I can't see the point in cheating at anything, but other's clearly do.
I'm not so much worried that AI is going to take over the human race but the ever increasing reliance on tech is worrying. I've just had to have my phone changed to digital. You know those times in a power cut where you could still call someone from your landline to check they're ok? That's going to be obsolete soon.
We don't appear to have any back up for when tech fails. I think I've got £30 cash because I just use my card or phone to pay for stuff. Some of you may remember that in 1987 there was a hurricane in Sussex & a lot of power cuts. Because we still had cash, we could go to the local shop to buy candles, torches & batteries etc could still get bread, milk & food. If we have a sustained power cut these days, some of us are going to be a little bit stuffed.
There needs to be a balance to work out what AI should & shouldn't be able to do. In my opinion it's moving so fast at the moment that it's too late to even attempt to put the brakes on. There's even a scam doing the rounds at the moment of the 'verify you're not a bot' thing on websites. Instead of getting you to click traffic lights they get you to open the 'run' menu on your pc by pressing keys & then you accidentally install their malware. I just want it all to roll back a bit.
Sorry for the essay & I may have slightly gone off on tangents but I suppose it proves I'm not an AI bot.
I tried the Southern Water chatbot the other day expecting it to be able to give the answer to a relatively simple question, nope I still had to ring them. Customer service chatbots have done a good thing for the local economy though in a very small way. I've had to buy 4 major appliances over the last 5 years, 1st was from Argos & trying to get out of the chatbot cycle to speak to an agent to say that my washing machine smells of burning was ridiculous. Next one Curry's & the dishwasher that they'd just installed was leaking, same thing. I browsed further than the 'big retailers' for my oven & spoke to a human at Carters they gave me sensible advice, fitted it perfectly & I had an 01273 number to call. I've just bought a gas hob from them, it's not been fitted yet as I don't want it yet. Again I was able to speak to someone about potential issues with sizing & when it's fitted I'm not going to worry that they don't know what they're doing. Curry's couldn't fit a dishwasher correctly I'm not going to trust them with gas! How is AI going to fit my cooker or dishwasher? How is it going to get it down my steps? How is it going to know the plumbing 'quirks' where I live?
I've also seen someone cheat their way through a degree with AI. They started using AI from the moment they started so progress in their essay standards have improved in line with their 'learning' from tutors & AI's learning. The degree topic is not something I would say was suitable for AI, I found it really bizarre why someone would want to cheat so much on something that is based on theories, rather than learn what the actual theories were & reach their own conclusion. To be fair I can't see the point in cheating at anything, but other's clearly do.
I'm not so much worried that AI is going to take over the human race but the ever increasing reliance on tech is worrying. I've just had to have my phone changed to digital. You know those times in a power cut where you could still call someone from your landline to check they're ok? That's going to be obsolete soon.
We don't appear to have any back up for when tech fails. I think I've got £30 cash because I just use my card or phone to pay for stuff. Some of you may remember that in 1987 there was a hurricane in Sussex & a lot of power cuts. Because we still had cash, we could go to the local shop to buy candles, torches & batteries etc could still get bread, milk & food. If we have a sustained power cut these days, some of us are going to be a little bit stuffed.
There needs to be a balance to work out what AI should & shouldn't be able to do. In my opinion it's moving so fast at the moment that it's too late to even attempt to put the brakes on. There's even a scam doing the rounds at the moment of the 'verify you're not a bot' thing on websites. Instead of getting you to click traffic lights they get you to open the 'run' menu on your pc by pressing keys & then you accidentally install their malware. I just want it all to roll back a bit.
Sorry for the essay & I may have slightly gone off on tangents but I suppose it proves I'm not an AI bot.
Normal Rob
Well-known member
To all those saying they don't like AI, then as a media/journalism professional of 30 years, please pay for the content you consume to help us continue to provide bias free, researched, objective journalism. It's that simple.
Frightening……. Good to watch though.
If it's IN ADDITION TO human interaction then great - it can be a great source of efficiency. But not if it's INSTEAD OF.
As an example, I had a rash appear one day (not anywhere embarrassing) and that evening I uploaded a photo of it to an AI website and got a diagnosis in 3 seconds. Next day I booked a GP appt which confirmed that the AI diagnosis was correct.
But I don't like it that photos/videos can appear to be real when they are not. I would like them to be flagged as being computer generated. I'm thinking of the adverts with hang-gliding goats and walruses who drive speedboats.
As an example, I had a rash appear one day (not anywhere embarrassing) and that evening I uploaded a photo of it to an AI website and got a diagnosis in 3 seconds. Next day I booked a GP appt which confirmed that the AI diagnosis was correct.
But I don't like it that photos/videos can appear to be real when they are not. I would like them to be flagged as being computer generated. I'm thinking of the adverts with hang-gliding goats and walruses who drive speedboats.
Knocky's Nose
Mon nez est retiré.
It's coming, wether you like it or not. The genie is out of the bottle...
Blockchain and Digital ID will straighten out the initial 'Wild West' - but it's going to be a bumpy road for now.
Blockchain and Digital ID will straighten out the initial 'Wild West' - but it's going to be a bumpy road for now.
WSU Dilettante
Active member
If anyone is interested Origin Story just did a really interesting 2 part podcast on AI (links below).
Part 1 is all about the history of AI, where it first came from, how it's generally used today.
Part 2 is more about the future, people's concerns, pitfalls, possibilities and so on.
In some ways AI is really going to improve our standard of living. It can read and interpret massive amounts of data that we couldn't even begin to comprehend. Its application in things like medicine alone are worth exploring, we just have to be careful to not exploit the human element. Theres been a lot of uproar recently about a podcast with an AI Michael Parkinson interviewing people, and there's already been strikes in Hollywood to safeguard actors and writers against the use of AI by studios.
But as highlighted in another post with the anagrams, it's got a long way to go and it's only going to be as good as its input and how well it's been trained. There was a twitter account that was run by and AI and within a few hours of creation it was basically a Neo-Nazi. Says a lot about Twitter!
Turns out we're training AI all the time, those annoying 'are you a human?' CAPTCHA things that we have to do sometimes, when we answer those we are actually training AI because it's still not smart enough to recognise the objects in the context of those images (think driverless cars).
Whether you're for or against is, it's all rather fascinating.
Part 1
Part 2
Part 1 is all about the history of AI, where it first came from, how it's generally used today.
Part 2 is more about the future, people's concerns, pitfalls, possibilities and so on.
In some ways AI is really going to improve our standard of living. It can read and interpret massive amounts of data that we couldn't even begin to comprehend. Its application in things like medicine alone are worth exploring, we just have to be careful to not exploit the human element. Theres been a lot of uproar recently about a podcast with an AI Michael Parkinson interviewing people, and there's already been strikes in Hollywood to safeguard actors and writers against the use of AI by studios.
But as highlighted in another post with the anagrams, it's got a long way to go and it's only going to be as good as its input and how well it's been trained. There was a twitter account that was run by and AI and within a few hours of creation it was basically a Neo-Nazi. Says a lot about Twitter!
Turns out we're training AI all the time, those annoying 'are you a human?' CAPTCHA things that we have to do sometimes, when we answer those we are actually training AI because it's still not smart enough to recognise the objects in the context of those images (think driverless cars).
Whether you're for or against is, it's all rather fascinating.
Part 1
Part 2