• Due to ongoing issues caused by Search, it has been temporarily disabled
  • Please excuse the mess, we're moving the furniture and restructuring the forum categories
  • You may need to edit your signatures.

    When we moved to Xenfora some of the signature options didn't come over. In the old software signatures were limited by a character limit, on Xenfora there are more options and there is a character number and number of lines limit. I've set maximum number of lines to 4 and unlimited characters.

Jobs That AI Will Replace. Very Soon!

That's why my real theory is that (on a timeframe worth discussing here) the Rembrants and the Hemingways and the Marlon Brandos of the world don't have anything to worry about.

The journeyman artists; the guy who writes commercial jingles, the dude who does background art for LEGO sets, the attractive but down to Earth mom on the Yogurt Commercial, their days might be numbered.
 
Plus, half the time, GPs are simply talking to people. Meeting a human is important for most patients. However, an AI system could chew over symptoms and test results and spot cases of diseases that the ordinary seldom meets and thus may not think of.

Hans


Except that, at least around here, we already don't have enough GPs to meet the demand. Current doctors are burning out and leaving the profession, and our archaic training program isn't meeting the demand for new doctors.

An AI that I can get an appointment with is far better than not having a regular doctor at all.
 
I don't see AI as replacing creative work, but being a tool people can use creatively when they lack the actual skills to implement their vision.



No one look at my avatar!

But yes, this is the way. "Creativity" doesn't necessarily go hand-in-hand with technical competence. Sure, the AI might be able to make a picture with the "marmot flipping burgers" et al., but would it ever conceive of the idea of burger-flipping marmots in the first place?
 
As my local GP does minor surgery (removing skin-tags and so on), innoculations, takes blood for blood-tests, does (necessary) hands-on examinations, I don't see GPs being replaced in the near future.

Plus, half the time, GPs are simply talking to people. Meeting a human is important for most patients. However, an AI system could chew over symptoms and test results and spot cases of diseases that the ordinary seldom meets and thus may not think of.

Hans

Agree with both of these. Not sure I would be trusting an AI to check my prostate.
 
Except that, at least around here, we already don't have enough GPs to meet the demand. Current doctors are burning out and leaving the profession, and our archaic training program isn't meeting the demand for new doctors.

An AI that I can get an appointment with is far better than not having a regular doctor at all.

I can easily envisage AI as the entry level to a GP and even specialists. You list your symptoms, initial tests are made, and at this point a fair portion of patients will simply get a prescription. Only the more complex cases, the surgical cases, or the ones with a significant psychological factor are sent on to a human doctor. .. And some are actually sent to some kind of therapist, instead.

Hans
 
I can easily envisage AI as the entry level to a GP and even specialists. You list your symptoms, initial tests are made, and at this point a fair portion of patients will simply get a prescription. Only the more complex cases, the surgical cases, or the ones with a significant psychological factor are sent on to a human doctor. .. And some are actually sent to some kind of therapist, instead.

Someone at my workplace is actually attempting to set up something along these lines. Not AI, but wants to get "all the symptoms" of every patient visit in the database, then use those to steer the patient to particular doctors "who treat those symptoms". It's not going to work because "symptoms" aren't recorded at all -- patient "reason for visit" isn't symptoms, and diagnoses aren't symptoms, and I won't even get started on the madness of attempting to connect a provider's board certifications to a patient's visit. Healthcare is more complicated than people think. Healthcare data is much more complicated than people think. However complicated you think it is, multiply it. Then multiply it again. And once more. It's never as easy as it sounds from outside.
 
Even simple things like reminders no longer being a sterile "You have a reminder to take your tablets" it becomes "Marjorie don't forget in half an hour you'll need to take your heart tablets" and "That TV programme you liked about the chimpanzee, there's a new episode on tomorrow, should we watch that?"

Good idea, and very positive. Would also be able to summon help when needed.

Depends on what the worker actually does. Do you think "office worker" is a single job? What do you think people do in offices? I've held several different jobs in offices, doing very different things. Some of those things could be automated, perhaps, but not all of them.

I'm using the generic term because there are so many of them. Accountants will redundant, legal executives likewise. Policy analysts, help desk and order-takers... I couldn't even begin to list all of the types of jobs that can be replaced by AI.

I feel like some are thinking "AI = Data from Star Trek" which may be true one day in the future, but right now AI doesn't even match the ship's computer on Star Trek and that computer was quite stupid. Couldn't even figure out what temperature to serve drinking water or memorize Picard's normal tea order!

I'm thinking more a computer doing it all rather than robotics. Humanoid robots' big opportunity is in age care, where the number of old farts needing care is far outstripping the ability to find people to wipe arses for minimum wage.

The other thing is, they don't need to be smarter or faster than a human except at one task. Thing of a robotic welder - it's useless for welding steel pipes in a factory, but 100 times more productive at manufacturing petrol tanks than a human.
 
Someone at my workplace is actually attempting to set up something along these lines. Not AI, but wants to get "all the symptoms" of every patient visit in the database, then use those to steer the patient to particular doctors "who treat those symptoms". It's not going to work because "symptoms" aren't recorded at all -- patient "reason for visit" isn't symptoms, and diagnoses aren't symptoms, and I won't even get started on the madness of attempting to connect a provider's board certifications to a patient's visit. Healthcare is more complicated than people think. Healthcare data is much more complicated than people think. However complicated you think it is, multiply it. Then multiply it again. And once more. It's never as easy as it sounds from outside.

Oh, I quite agree. Just like battle-plans, no computer application survives the first confrontation with reality. ... Which is really a good thing.

Hans
 
They've also looking at automating Bias response since 2020...


Studies have shown that implicit bias—the automatic, and often unintentional, associations people have in their minds about groups of people—is ubiquitous in the workplace, and can hurt not just employees, but also a company’s bottom line.

For example, employees who perceive bias are nearly three times as likely to be disengaged at work, and the cost of disengagement to employers isn’t cheap—to the tune of $450 billion to $550 billion a year. Despite the growing adoption of implicit bias training, some in the field of human resources have raised doubts about its effectiveness in improving diversity and inclusion within organizations.

But what if a smart device, similar to the Amazon Alexa, could tell when your boss inadvertently left a female colleague out of an important decision, or made her feel that her perspective wasn’t valued?


https://news.northeastern.edu/2020/01/29/how-about-a-smart-device-that-could-catch-implicit-bias-in-the-workplace/
 
https://creator.nightcafe.studio/creation/FhiAjB6R2xsww3QxDIf0

(Sadly I hit the character limit and could only get about 1/4th of the prompt actually in there.)

ETA: I assumed it was obvious but just to check all the marks in accordance with the new rules, the link is to an AI created image from the NightCafe AI.
It's not entirely clear, but I think the trunk-generating rules need tweaking.

Of course you never know how long it will take for this stuff to be perfected. My vote is that it will be driven by porn, and happen faster than you expect.
 
I guess they are good for explaining what is wrong with gamers in a bitchy female voice from what I have heard.

I have read a few responses and now realize this is not where the thread is going. I was looking up ai girlfriend and it sounds not ready for prime time
 
Last edited:
Depends on what the worker actually does. Do you think "office worker" is a single job? What do you think people do in offices? I've held several different jobs in offices, doing very different things. Some of those things could be automated, perhaps, but not all of them.

I feel like some are thinking "AI = Data from Star Trek" which may be true one day in the future, but right now AI doesn't even match the ship's computer on Star Trek and that computer was quite stupid. Couldn't even figure out what temperature to serve drinking water or memorize Picard's normal tea order!

I always saw that one as Picard's inability to use the technology...

The only thing he could get out of replicators was 'Earl Grey Hot', so he relied on turning up at crew member's quarters at meal times, or ate at ten forward or one of the messes.

Painted a portrait of an aging Captain who was completely out of his depth with all the technology around him...

The crew were fond of him in a 'that old uncle' kind of way, so did their best to carry him...
 
Someone at my workplace is actually attempting to set up something along these lines. Not AI, but wants to get "all the symptoms" of every patient visit in the database, then use those to steer the patient to particular doctors "who treat those symptoms". It's not going to work because "symptoms" aren't recorded at all -- patient "reason for visit" isn't symptoms, and diagnoses aren't symptoms, and I won't even get started on the madness of attempting to connect a provider's board certifications to a patient's visit. Healthcare is more complicated than people think. Healthcare data is much more complicated than people think. However complicated you think it is, multiply it. Then multiply it again. And once more. It's never as easy as it sounds from outside.

A computer program to diagnose a patient's issues could work, but only if it is done right. Doctors would have to give heaps of feedback before it would work in reality. Once it does work, it would be enormously powerful.
 
A computer program to diagnose a patient's issues could work, but only if it is done right. Doctors would have to give heaps of feedback before it would work in reality. Once it does work, it would be enormously powerful.
I don't see why doctors would need to give heaps of feedback. Clinical trials would be more important. There's already a fair amount of diagnostic software (AI and otherwise) in the world. Some researchers at MIT have recently developed a neural network capable of early detection of Parkinson's from analysis of breathing patterns. Notably, this isn't replacing anybody or anything; it's providing a diagnostic service that could not be done at all before.

....

I do find it funny that in the movie Her, from all of ten years ago, Joaquin Phoenix's near-future job is whore-poet: he writes nice personalized notes on behalf of other people in greeting cards, a job that I suppose somebody thought would be AI-proof, but in reality would already be on the verge of extinction if there had ever been a market for it in the first place.

Anyway, when I was a kid I kind of dreaded a world where strong AI would replace us. Now I consider that preferable to the dumbtopia where bedazzled technophiles insisted that weak AI should.
 
I can easily envisage AI as the entry level to a GP and even specialists. You list your symptoms, initial tests are made, and at this point a fair portion of patients will simply get a prescription. Only the more complex cases, the surgical cases, or the ones with a significant psychological factor are sent on to a human doctor. .. And some are actually sent to some kind of therapist, instead.

Hans

Maybe, but one of the current failures of the UK GP system is that GPs don't have the time to talk to, and reassure, people who may be unwell, may be lonely and so on. "Old fashioned" GPs had the time to do this (presumably because lifespans were shorter in old fashioned times). My late father was a "frequent flier" at the local GP surgery because he was the perfect storm of loneliness and lifelong hypochondria.

Unfortunately AI may not be able to help with this too much. These patients want to speak, at length, to a human - which may be one reason why alternative therapists are so popular.
 
I think there will be new jobs, but some people will necessarily get shafted just as the Luddites did. Retraining sounds good on paper, but an agoraphobe will be hard pressed to succeed in a "people job" if that's suddenly the only game in town.
 
This can go here.
 

Attachments

  • 727FF2EB-1E52-4884-9F26-7FD629E774C9.jpg
    727FF2EB-1E52-4884-9F26-7FD629E774C9.jpg
    54.4 KB · Views: 10
Bill Gates today on AI:

No matter what, the subject of AIs will dominate the public discussion for the foreseeable future. I want to suggest three principles that should guide that conversation.

First, we should try to balance fears about the downsides of AI—which are understandable and valid—with its ability to improve people’s lives. To make the most of this remarkable new technology, we’ll need to both guard against the risks and spread the benefits to as many people as possible.

Second, market forces won’t naturally produce AI products and services that help the poorest. The opposite is more likely. With reliable funding and the right policies, governments and philanthropy can ensure that AIs are used to reduce inequity. Just as the world needs its brightest people focused on its biggest problems, we will need to focus the world’s best AIs on its biggest problems.

Although we shouldn’t wait for this to happen, it’s interesting to think about whether artificial intelligence would ever identify inequity and try to reduce it. Do you need to have a sense of morality in order to see inequity, or would a purely rational AI also see it? If it did recognize inequity, what would it suggest that we do about it?

Finally, we should keep in mind that we’re only at the beginning of what AI can accomplish. Whatever limitations it has today will be gone before we know it.

https://www.gatesnotes.com/The-Age-of-AI-Has-Begun
 
Maybe, but one of the current failures of the UK GP system is that GPs don't have the time to talk to, and reassure, people who may be unwell, may be lonely and so on. "Old fashioned" GPs had the time to do this (presumably because lifespans were shorter in old fashioned times). My late father was a "frequent flier" at the local GP surgery because he was the perfect storm of loneliness and lifelong hypochondria.

Unfortunately AI may not be able to help with this too much. These patients want to speak, at length, to a human - which may be one reason why alternative therapists are so popular.

I'd challenge that - I think they'd be happy with chatting to anything* that responded to them with sympathy and support. I really do think the ability to "chat" with people is an area where AI can make immense improvements in people's life without "doing" anything.


*Look at how many pet owners talk to their animals and even have conversations with them!
 
Okay fair point. Fear of being replaced by new tech hasn't exactly born out historically speaking.

I'll amend that to people who mocked the fears (justified or not) of being replaced by machines when it was blue collar factory workers and truckers now having a different reaction when it's ARTISTS! (Flourish) and actors and writers possibly being replaced is, again not universal but a subtext in a lot of this.

Also several IT roles, especially in the Service Delivery area. As a guy I respect said "We've done computerization. Now we're doing computerization of computerization" and a lot of the people in nice safe "I'll reject changes like my manager and his manager before him" jobs are in for bad news.
And yeah, my first job in 1984 someone told me my job would be obsolete in a couple of years as "magic bullet product" would do it all. Nope, now I had the equivalent of a lawn mower instead of a scythe so I could do my job and look at doing it better and other stuff. And on and on.
 
Perfect job for AI - company for the lonely. The number of times I've called elderly people and have been on the call for an hour or so is staggering. All they want is a chat.
 
Two of the biggest white-collar sectors who should be studying for their barista qualifications right now are those at banks and insurance companies.
I can't think of a worse industry to be in right now. Humans shouldn't be used as glorified adding machines.

Engineering, too, will be changed out of all recognition. Who needs 20 traffic engineers, when you have one computer doing the same job much fast and better than the team of professionals costing $200k each?
"Hey ChatGPT, your bad planning caused a 20% increase in traffic accidents. Prepare to be sued for $100 billion!"

How do you think traffic engineers currently do the job, with protractors and slide rules?

Given our slavish devotion to the Friedman Doctrine, companies will be falling over themselves to get rid of people and invest in machines.
I'm not so sure about that. In this day and age, why do we still have skyscrapers full of offices?

Advertising agencies? Who needs 'em? A single AI computer will devise a campaign that works without human input. We know how easy it is to persuade people, thanks to my pal Robert Mercer and Cambridge Analytica.
Advertising still needs creativity to be more successful than the competition. AI is a long way from replacing humans there.

The next ten years are going to see a complete change in employment dynamics, and white-collar workers are going to be a dying breed.
One can hope, but I doubt it.

OpenAI’s ChatGPT Reportedly Costs $100,000 a Day to Run

I could employ over 100 traffic engineers for the price of running one AI that is useless for the job. I would probably need 100 traffic engineers just to train it, another 100 to check that it was doing the job properly, and 100 lawyers to handle all the torts. And they would all want more money for dealing with an AI.
 
Perfect job for AI - company for the lonely. The number of times I've called elderly people and have been on the call for an hour or so is staggering. All they want is a chat.

They should join forums like this one. There are many people who would be willing to discuss almost any topic.
 
Perfect job for AI - company for the lonely. The number of times I've called elderly people and have been on the call for an hour or so is staggering. All they want is a chat.

Early on I saw a report on Alexa from parents of autistic children who noted that she was an excellent companion for them as she did not mind if she was asked the same question over and over and yet show no frustration.

As an elderly person (but not hugely lonely) I just asked Alexa "If I was lonely would she tell me a story? She replied that my text to speech app was not loaded and I had no books in my library. Then she suggested on the screen a joke and that I ask, "Why beavers are the best story tellers?". I did but am not telling.
 
Yeah but socially that is super problematic.

People can already get like half of Maslow's Hierarchy of Needs with a 200 dollar Chromebook and a 50 dollar a month internet connections. What happens when they can get all of them?
 
Yeah but socially that is super problematic.

People can already get like half of Maslow's Hierarchy of Needs with a 200 dollar Chromebook and a 50 dollar a month internet connections. What happens when they can get all of them?

What?

Physical- not provided by pc
Safety- not provided by pc
Love- not provided by pc (unless you're into that)
Self Esteem- not provided by pc
Self actualization- not provided by pc.

The best you could hope for is a heavily watered down simulation of a couple of them, or a totally ineffective comparison, like saying safety is provided by ordering a security system online.

Plus if you spend $200 for a Chromebook, you got ripped.
 
Yeah but socially that is super problematic.

People can already get like half of Maslow's Hierarchy of Needs with a 200 dollar Chromebook and a 50 dollar a month internet connections. What happens when they can get all of them?

But we know we have (at least in the UK) a problem with people who are unable to regularly have social interactions. The problem of loneliness especially in housebound folk is endemic.

I really do see that a "chatty" AI assistant device that responds like a person - albeit always a level headed person - that can initiate conversations could be very useful in helping folk who are lonely. We recommend getting a pet for people to help with loneliness so I don't think alleviating a problem like loneliness and lack of social interactions requires other people.
 
Healthcare is more complicated than people think. Healthcare data is much more complicated than people think. However complicated you think it is, multiply it. Then multiply it again. And once more. It's never as easy as it sounds from outside.
True, but a system that helps you perform an accurate self-diagnosis would go a long way towards reducing pressure on doctors.

You don't want an AI for that though. Like a doctor, it just puts a middleman between you and what you need to know.

One application that an AI would be ideal for is scamming. You think you are talking to a real person, but it's actually just an algorithm designed to clean out your bank account. Lonely old people would be ideal marks.
 
Someone at my workplace is actually attempting to set up something along these lines. Not AI, but wants to get "all the symptoms" of every patient visit in the database, then use those to steer the patient to particular doctors "who treat those symptoms". It's not going to work because "symptoms" aren't recorded at all -- patient "reason for visit" isn't symptoms, and diagnoses aren't symptoms, and I won't even get started on the madness of attempting to connect a provider's board certifications to a patient's visit. Healthcare is more complicated than people think. Healthcare data is much more complicated than people think. However complicated you think it is, multiply it. Then multiply it again. And once more. It's never as easy as it sounds from outside.

If I bring up HL7, FHIR, SNOMED CT and stuff do you promise not to scream or mutilate me?
 
But we know we have (at least in the UK) a problem with people who are unable to regularly have social interactions. The problem of loneliness especially in housebound folk is endemic.

I really do see that a "chatty" AI assistant device that responds like a person - albeit always a level headed person - that can initiate conversations could be very useful in helping folk who are lonely. We recommend getting a pet for people to help with loneliness so I don't think alleviating a problem like loneliness and lack of social interactions requires other people.

Don't know about the current octogenarians, but if I'm ever old enough to have someone try to "solve" my problems by having a non-sentient machine talk to me, I'm throwing the glorified AIBO into their face.
 
"Wine label copywriter" is an actual job title. I discovered that just the other day.

Here I thought AI had long ago taken over writing the incoherent bilge that I try not to waste neurons reading when I handle a bottle o' plonk. Surely it could be done?

And wine & food criticism while the robot's at it.
 
Interesting point about art and technology: When photography was introduced, some people predicted it would be the end of painting art. Not only was it not the end of painting art, it wasn't even the end of figurative art, AND photography turned into an artform of its own.

Hans

There was a vogue in early photography for a painterly look, as far as B & W would go. Genre photography of that type can be quite impressionistic. With applied color (another early vogue; it's come back into fashion now, of course) it could -- can -- fool you into thinking it's painting.
 
Last post in this thread was almost exactly one year ago.

Another year, another study points to an impending "job apocalypse" thanks to our robot friends.

AI could eliminate nearly 8 million jobs in UK, study shows

The technology could herald a "job apocalypse," the study said.

That's out of about 36 million jobs currently held by someone in the UK.

Researchers analyzed 22,000 tasks carried out by workers across the U.K. economy, finding that 11% are currently exposed to the threat of displacement by AI, the study said. The jobs at greatest risk include entry-level, part-time and administrative roles -- a set of positions disproportionately held by women, the study added.

"Tasks" and "jobs" aren't quite the same, are they? There are a lot of "tasks" that I used to perform in the course of doing my job that I no longer need to perform, or that have been made orders of magnitude easier thanks to technology.

The report describes a soon-to-begin phase of AI adoption during which some of these “low-hanging fruit” jobs will be replaced by the technology. The overall workforce impact over the period could be limited, the study said, but some roles will experience massive effects, such as the elimination of one-third of administrative jobs.

A second phase could bring much deeper integration of AI that will threaten up to 59% of tasks, the report said. If companies allow AI to access proprietary information and execute key tasks, the study said, the resulting disruption may slash a wider swathe of jobs, including a larger share of high-paying positions.

I will note that the actual number of jobs seems to be at a record high, so it hasn't happened yet.
 
Back
Top Bottom