HALF the companies using ChatGPT say it has already replaced workers (Fortune)

[highplainsdem]

https://fortune.com/2023/02/25/companies-replacing-workers-chatgpt-ai/
Archive link: https://archive.ph/9uZly

OpenAI initially launched ChatGPT toward the end of November, but the artificial intelligence chatbot had its stable release in early February. Earlier this month, job advice platform Resumebuilder.com surveyed 1,000 business leaders who either use or plan to use ChatGPT. It found that nearly half of their companies have implemented the chatbot. And roughly half of this cohort say ChatGPT has already replaced workers at their companies.

-snip-

Business leaders already using ChatGPT told ResumeBuilders.com say their companies already use ChatGPT for a variety of reasons, including 66% for writing code, 58% for copywriting and content creation, 57% for customer support, and 52% for meeting summaries and other documents.

In the hiring process, 77% of companies using ChatGPT say they use it to help write job descriptions, 66% to draft interview requisitions, and 65% to respond to applications.

-snip-

As gung ho as business leaders appear to be about the potential of ChatGPT, it’s not without its critiques, including concerns regarding cheating and plagiarism, racism and sexism bias, accuracy, and overall questions about how it’s been trained to learn. The Atlantic’s Ian Bogost warned it should be treated as a toy not a tool, and New York Times tech columnist Kevin Roose said that Microsoft’s new A.I. version of its Bing search engine powered by ChatGPT’s OpenAI left him feeling “deeply unsettled” and “even frightened” after a two-hour chat in which it sounded unhinged and somewhat dark.

-snip-

The article then reminds readers that OpenAI CEO Sam Altman has warned that ChatGPT shouldn't be relied on for anything important.

But hey, why should top execs care about accuracy or plagiarism, or hallucinating AI that just makes stuff up, when they can cut payroll?

More survey findings:

Nearly 100% of the companies say ChatGPT has saved them money. Nearly half say it's saved them over $50,000. Just over 10% say it's saved them over $100,000.

You can save a lot of money firing people and replacing them with AI.

More than 90% of the companies say they plan to expand their use of AI.

So more jobs will be lost.

Welcome to the AI future.

Try telling me again how wonderful ChatGPT is for society.

[edisdead]

1. Have people not been paying attention for the last 20 years?

How is this a surprise to anyone? Hell, heading into 7th grade way back in whatever year in the 80’s that was we were talking about this and why it was going ti be super important for kids of that age to understand computers, and the factory workers started getting replaced by robots, and the robots started beating people at chess and then we started making software that helped artists and musicians and could improve their scores, enhance their photography and every other thing that technology helps with….

How is this a surprise? Tech has replaced workers forever and that isn’t ever going to change. It is going to be more impactful and we need to start figuring out how that effects the economy and what life is like without the jobs and careers that are going to be replaced. But it sure as hell shouldn’t be a surprise to anyone.

[highplainsdem]

3. What part of "the OpenAI CEO says not to rely on ChatGPT for anything important"

don't you understand?

It's a DEFECTIVE product.

It should never have been released.

It got released while still defective because all these companies wanted to be first.

But all the execs firing people can see is a chance to slash payroll.

So we're going to see huge social and economic disruptions because of the mad rush to implement defective technology.

And everyone foolish enough to add to the hype shares some of the blame.

[Silent3]

6. The poster you just responded to is talking about AI in general, not ChatGPT in particular

Sooner or later (hopefully using much better tech than ChatGPT) AI is going to replace a lot of human workers. That's the point I think the poster was trying to make.

There's really no practical way to avoid this. No one will ever outlaw this entire area of technology to protect human jobs. We have to figure out how to cope with it.

In the meantime, I sure hope we don't get a lot of human workers replaced by shitty AI doing a shitty job of whatever the humans used to do.

[highplainsdem]

7. Thanks, but I was aware of that, just as I'm aware of tech replacing jobs.

My OP is about the insane rush to use ChatGPT when it's defective, and fire people asap.

I've posted, too, about Sam Altman and others saying AI needs to be regulated. But they aren't giving governments enough time to regulate it.

I'm stunned by the willingness to release defective products, and products violating copyright, and products that will make teaching much more diffucult. They're creating a disaster.

Btw, Microsoft knew its Bing.AI would go haywire, in the same way it did when they released it a few weeks ago and then had to throttle it to limit the psycho responses. They knew because of a trial a few months ago in India where the same thing happened. They gave it a wider release anyway. It's all about money and beating the competition.

[edisdead]

60. Oh there have been many people rushing to use buggy software.

That’s happened many times.

And products in general not necessarily tech have been in wide use while not being perfected. (I am looking at you automobile)

I do not see how you stop this from happening.

[highplainsdem]

8. See this OP I just posted for ChatGPT's problem with accuracy:

https://www.democraticunderground.com/100217681596

[Silent3]

10. Oh, I've heard all about that

ChatGPT definitely isn't ready for primetime. It's remarkable at aping plausibly-human sounding text. It's terrible at accuracy.

[highplainsdem]

11. But there's a stampede to use it anyway. No one would have

rushed to use calculators if they'd often given inaccurate, sometimes wildly inaccurate, results. No one would have used early word processors, stand alones or word processing software, if they'd printed the wrong words, lines of gibberish, or sentences that hadn't been typed but instead had been hallucinated by the word processor. I don't understand the tolerance of ChatGPT's mistakes and hallucinations. Fact-checking and proofreading can take a lot of time, and I've seen a lot of articles recommending that everything ChatGPT produces should be fact-checked. Do businesses think they can just excuse important errors as a chatbot glitch, and not be held accountable?

[edisdead]

59. This is exactly my point. And thank you.

[edisdead]

15. Did I limit my point to openAI?

Nope, my point is that this is happening whether or not openAI is a flawed product right now or not.

[highplainsdem]

17. Then thank you for posting an irrelevant reply.

[edisdead]

19. Are you just looking to argue?

So strange.

My point is (as another poster’s in this thread) is a general reply about technology and how we have been heading here for some time. I would argue since the wheel and fire. This will not be the last technology that people rail against. Not sure why that seems to upset you.

[highplainsdem]

20. You posted in this thread to argue, as you have in other threads I've posted about AI.

[edisdead]

21. In general I don't keep tabs on who I reply to.

I just don’t care enough about who I am replying to as much as the content. I do find it interesting that you do and might offer that because you do you felt the need to be argumentative. I could be wrong.

My intent wasn’t to invite argument, but again to illustrate the larger point that this is simply tech expanding. I believe there is a lot of hysteria surrounding it right now (not saying you are hysteric about it) and with that we end up with irrational responses and possibly bad law.

[dalton99a]

2. "I'll have my chat agent talk to your chat agent"

It will work wonders on dating apps

[Captain Zero]

4. Should smooth out a lot of bumps.

For certain.

[highplainsdem]

5. Re dating apps and ChatGPT, see this article:

https://www.independent.co.uk/life-style/chatgpt-ai-ios-dating-app-b2283598.html

[dalton99a]

9. +1. Damn

[Renew Deal]

32. That was funny.

Thanks for sharing. The first message in the article sounded like a sales pitch. I guess dating apps are marketing and sales pitch but it is obviously inauthentic.

[progree]

12. Some top execs, low level supervisors, and everyone in between don't care about quality

or accuracy.

But hey, why should top execs care about accuracy or plagiarism, or hallucinating AI that just makes stuff up, when they can cut payroll?

Once upon a time, I was a new engineer. As a brand new engineer needing something to do, my boss often gave me unimportant projects that nevertheless some kind of response was needed. It was get it done quick, if they (the recipient) complains, well, no problem here.

A chatGPT type of solution that just spewed out something that somewhat fit the request would have been just fine with him/them. Perfect actually.

A person like me and most engineers who take pride in their work and just can't knowingly and deliberately produce crap -- that wasn't a fit at all.

By the way that boss, even though an electrical engineering superintendant at a major electric utility, ended up leaving and working for his father's auto dealership. I had forgotten about him all these past 40 years. Then about 5 years ago I read in the paper that he was responsible for cheating many law enforcement agencies on equipment upgrades to their vehicles -- hundreds of vehicles. It blew me away to see his name (and I verified that yup, that's the same guy for sure). And yes, he pressured employees to do this stuff and one or two of them eventually whistle-blew on him.

[highplainsdem]

28. What a sad story. Sorry you had to work for anyone like that.

A lot of people are likely to be hurt by AI being misused by people in the business world who are like your former boss. But some are going to be hurt just by mistakes made by AI that people thought they could catch, but didn't.

We need regulations now, with companies yanking defective or legally questionable AI. We have the opposite.

[IcyPeas]

13. By 2029, computers will have human-level intelligence

By 2029, computers will have human-level intelligence,” Kurzweil said in an interview with SXSW

Ray Kurzweil, Google's Director of Engineering, is a well-known futurist with a high-hitting track record for accurate predictions.

https://futurism.com/kurzweil-claims-that-the-singularity-will-happen-by-2045

[Renew Deal]

14. A tale as old as time

Every one of these revolutions leads to changes in the way people work. From typewriters being replaced with PC’s to machines replacing hand crafting and even books replacing wise men as a source of information.

The funny thing about the example in the article is that it sounds like HR is replacing itself.

Generative AI is very good at creative writing. It still relies on humans to deal with accuracy of facts and meaning. People can work with that if they aware of it. It’s no different than learning to properly use other tools like hammers and hair brushes.

[edisdead]

16. Exactly

[highplainsdem]

18. I'm sorry you can't see the difference.

[Renew Deal]

33. Which difference?

Not sure what you mean.

[New Breed Leader]

69. bullshit

This is nothing like the revolutions that came before, this is on a much larger destructive scale.

[Renew Deal]

71. Care to explain?

We may actually agree, though warnings about doom due to technological changes is nothing new either. The one thing no one on DU is talking about is what happens if a society doesn’t need to work because of automation. Other sites including OpenAI’s Discord are talking about this.

[Kid Berwyn]

22. ChatCEOs

Then they might see the problem.

[highplainsdem]

29. More and more top execs will start losing their jobs.

There are people in AI talking about the need for a Universal Basic Income. But they have no way to ensure that is implemented as people are put out of work.

[Kid Berwyn]

37. We the People better get that sooner rather than later into law.

Intelligent Machines may be the only things more heartless than the typical CEO.

[honest.abe]

23. You are fighting a losing battle.

AI is here to stay and will have more and more influence in our daily lives.

[Elessar Zappa]

24. Exactly.

We gotta learn to live with it. There will be some negatives from AI but also many positives, especially in the field of medicine.

[honest.abe]

25. Its funny.. even one of the biggest complaints about students cheating...

Turns out many teachers are now using AI to help with their lesson plans.

They realize the benefits outweigh the negatives.

[highplainsdem]

27. Teachers doing that are lazy and/or stupid and/or incompetent. Just

as they'd be if they hired someone else to do it for them.

But yes, they have surrendered any right to complain about students cheating in the same way.

They can all be dumbed down AI-assisted morons together.

Who needs education when you can have AI-assisted pseudo-education?

Thank God some teachers and students still have integrity.

[honest.abe]

48. That's an absurd statement.

You should apologize to all those hard working teachers simply trying to do the best they can to teach our children with limited resources.

[highplainsdem]

54. It isn't an absurd statement. I have the greatest respect for teachers who

take their profession seriously. A lot of my friends over the years have been teachers (ones I met as an adult, not while I was in school), from elementary school teachers to college professors (more of the later), and my extended family has included college professors and a high school teacher.

But not all teachers are good. Not all teachers are responsible. And not all teachers will try to make sure their students are learning rather than letting AI do the thinking and writing/creating for them.

This is one time students really need teachers to stand up for them, not cave to new technology.

I saw a column from a teacher who said his kids would just cheat with AI anyway, so he'd let them cheat because they'll be using AI when they graduate. He didn't want to be bothered trying to figure out if they were cheating. I pity his students.

Teachers should be able to write their own lesson plans, just as they should know what they're teaching.

And btw, a study I saw and posted about said a higher percentage of students than teachers felt ChatGPT should be banned, and even the kids using it considered it cheating. The students know better than some teachers on this subject.

[honest.abe]

56. How would that work.. "ChatGPT should be banned"?

Who, where and how would that be enforced? Sure you could block access in the school network but many have internet access through their cell phones. It seems a losing battle with technology winning. I think in the classroom situation, writing assignments and test taking should all be in the classroom with no cell phones. You know there is already a huge issue with students buying online help with writing papers and take home assignments long before ChatGPT. Do you not care about that??

This AI chat technology is actually still in its early stages. If you are upset about this now, you will go apeshit in a few more years.

[highplainsdem]

26. Awww... The"lie back and enjoy it" argument. Ukraine should surrender

at least some of its territory, since it can't win.

Don't try to regulate gun ownership and use.

Don't post speed limits.

Don't fight for civil rights.

Don't fight for workers' rights.

Why would anyone be a social justice warrior concerned about the harm done by AI?

Even the people promoting AI admit it has the potential to do great harm and should be regulated.

[Renew Deal]

34. I think the point is you're focusing on the harm part and not the regulation/education part.

I think you should change your approach from “look at this imperfect/bad thing AI did” to educating people on how to use it safely, suggesting workable regulations, highlighting ethics concerns and how to alleviate them, and highlighting privacy concerns with practical suggestions on dealing with them. You have a chance to be a leading thinker on this topic and it’s a much better message. It will also be more helpful.

[highplainsdem]

40. When I see articles on workable regulations and other approaches to mitigating the harm AI will

do, I'll post them. I did mention an approach one British school was planning to take, basically shifting class time to testing what the students will be expected to learn at home. But the teachers I've known didn't go into teaching to spend all or most of their time giving tests. And for that matter, you don't need a teacher for that testing, just someone to watch the kids so they don't cheat while in the classroom.

This is DU, and I'm not being paid to write here and don't have a large audience here. I can't become "a leading thinker" on this or any field posting here. I post here to entertain myself (mostly in the Lounge or Music Appreciation for that), to post info when I think it could be important for DUers to know about, to relay interesting opinions or comments I run across, to interact with people I like here, and sometimes to vent, though not about personal matters (though I'm fine with other DUers doing so if they want to).

I do wish I knew someone working in AI at a high level, because I would like to talk to them about the harm AI can do and the ethical issues, and possibly relay anything interesting they had to say to DU, even if anonymously. I noticed that when the editor of the science fiction magazine Clarkesworld had to close the magazine to submissions (which I'd predicted was likely to happen even before I'd heard about the problems at Clarkesworld), one of the people replying in his Twitter thread said they worked in AI but were really sorry to see this happening. They didn't elaborate and I didn't check back later to see if they'd said any more. But the only time I ever talked to anyone I'm sure had anything to do with AI was when a friend who'd been a student and was then a colleague of Marvin Minsky used Minsky as his answering service and gave me Minsky's office number, decades ago before cell phones when we were both moving but needed to stay in touch, and told me to talk to Minsky to leave messages and my new number as soon as I had it. I apologized to Minsky, who told me he didn't mind, but advances in technology beat using friends or relatives for answering services. I am not a Luddite and never have been.

Anyway, I'm going to keep posting the articles I run across on the harm done by AI. I've known too many teachers, musicians, producers, artists, writers and editors (of fiction and nonfiction), and actors and screenwriters over the years not to be concerned by the harm it's already doing, and angered by the naive assumption that people using AI to teach, create music or art, write or edit, or "replace" actors are really doing that themselves and deserve any respect.

If you want to complain about what I post, go ahead. But I have good reasons for posting what I do on the topics I post on.

Even cats. Though I'm cat-less at the moment.

[honest.abe]

49. You sound like a buggy whip salesman from a bygone era.

Your analogies are clearly irrelevant to the discussion. This is about a new technology that will improve our lives if used correctly.

[Lancero]

61. Oh hey, looks like we got a modern replacement for Reductio ad Hitlerum.

[area51]

30. All the more reason why we need

a universal basic income along with universal healthcare.

[highplainsdem]

31. Yes. Now tell me how we get there politically.

Companies happy to lay off workers to save money will fight having to pay higher taxes to support non-employees and keep them healthy.

And how high does UBI have to be to support a consumer-driven economy?

[Meowmee]

38. Yes

[dembotoz]

35. anytime you see the letters roi it means r i p.....someone lost a job

I work in technology

do lots of webinars, go to some conferences, see lots of demos.

I am old enough to be held in awe with what some of these new products can do.

We have become a self service economy.
We like that we can order with a click
jobs eliminated
self check out
jobs eliminated
check you bank stuff online
jobs eliminated
do pretty much anything on line
jobs eliminated.

[Meowmee]

36. Pretty soon we won't need people for anything anymore

Pretty soon we won’t need people for anything anymore… maybe in the long run it will help lower the population so, it could be a good thing… lol.

[Samrob]

39. Now we will blame Biden for job losses as we did blame Clinton's NAFTA when automation took jobs

We just never learn from history.

Right now, we would not have to be so afraid of China had we gone with Obama's TPP. And almost everything opponents said about TPP is in place and happening now WITHOUT TPP. Read about it at the many links around the Internet. We have little leverage for effective sanctions against Russia now.

[Prairie_Seagull]

41. Our nation was promised these tech. jobs. I am pretty sure it was a way to promise us that the-

outflow of jobs would not cause as much pain as it did. Burger, and other service jobs are numerous. Tech jobs are generally higher paying with benefits. Who benefits from all of us flipping burgers or.... In ten years we could mostly all be low wage slaves. This is going to be a lot worse if and before it gets any better

AI ties into what this country will be in the future. In not just my opinion. It is likely to have huge unintended consequences. It's my belief that we are going in to a new era/age. This is not a non issue. I believe this is going to have huge ramifications.

Many have said 'what could go wrong' implying that many things could go wrong. I would say 'are' already going wrong.

Thank you for keeping us aware HPD. You ROCK

We are all going to meet this beast.

[andym]

43. Robots powered by AI could replace the burger jobs too

if it turns out that they cost less than minimum wage over a year.

[highplainsdem]

45. That's what they're planning on.

[highplainsdem]

46. Thanks, Prairie_Seagull! Our society has to decide, and

very quickly, if it's okay with human workers and creators being replaced by AI.

Because unregulated AI is coming for all, or almost all, of the work people do to earn a living and give their lives purpose if they enjoy their work. And it's coming for the art and music and writing they create and enjoy.

And we have absolutely no reason to believe those profiting from AI will care about the workers losing their jobs, including the creative people being told AI can do as well or better, and more cheaply.

Netflix Japan has already had AI replacing some anime artists. The excuse given was that there was a labor shortage, but there was no evidence of that...and if there was one, it was due to anime artists being paid as little as $200 a month.

Even people working at jobs paying poverty wages will be replaced.

[andym]

42. ChatGPT is very good at writing copy or marketing material

Those jobs will be on the line. CNET has had it write articles for them. It is also surprisingly good at computer programming. It;s only matter of time when low-level programmers are replaced.

[highplainsdem]

44. See these threads for what happened with CNET and

another magazine:

https://www.democraticunderground.com/100217572530

https://www.democraticunderground.com/100217574806

https://www.democraticunderground.com/100217635463


AI makes mistakes with coding, too. It's still being used for coding, including for malware.

[JCMach1]

51. Precisely, human intervention is still required... In short, it only eliminates busy work

I will straight up tell you kids are using it to finish off busy work at school.

Same in the workplace.

However, having used and experimented with it extensively, writing still requires a human if it steps beyond the numbingly simple.

So yeah, great for real estate description...

Horrible at a personal analysis of a piece of literature.

[Emrys]

47. I copy-edit books for a living.

During my career, I've already had to deal with books by non-first-language English speakers that were passed from other languages through Google Translate etc., and nobody'd realized this until I came to work on them and sounded the alarm.

If instructed to press ahead, the work could be gruelling to produce something that was passable as publishable from often very primitive and frequently more than borderline-nonsensical "translations".

That work sometimes showed up plagiarisms from Wikipedia etc. I had to be careful in working on improving such translations to rephrase them so that translated passages from, say, the Korean version of Wikipedia were sufficiently different from the passages in the English-language Wikipedia entries to not be identifiable plagiarism.

These advances in AI will add whole new dimensions to similar problems for we copy-editors that I dont relish facing!

[highplainsdem]

62. What a mess! Good copy editors are a treasure, and from what I've heard, usually

not paid what they're worth.

I was reading a bit about a book today that was completely written by ChatGPT at the direction of a blue collar worker in an important and necessary profession. But he'd heard of someone else using AI to write a book. He even let ChatGPT choose the topic. And he self-published it on Kindle, and seems absurdly proud of his "achievement."

Pathetically, although he thought the ChatGPT-written novel borrowed a lot of ideas from one book, what bothered him more was it having the exact same title as another book (hell, anyone who's read much, let alone become familiar at all with publishing, knows how common that is). So he removed one word from the title. Which seems to have been his main creative contribution to it.

I'll give him props for at least being honest about ChatGPT having written it for him.

But he isn't a writer when ChatGPT did the writing for him by plagiarizing writing done by humans.

And the people trying to pass off ChatGPT-written stories, articles and books as their own aren't writers, either. They're frauds, and IMO contemptible.

And I hope for his sake he never finds his own safety dependent on someone supposedly in his profession but totally inexperienced and untrained, who asked ChatGPT for help fixing something crucial.

[JCMach1]

50. I don't buy it. ChatGPT is like a next step card catalog, to net, to ChatGPT

It eliminates simple busy work, that's all.

Remember typing pools? It's the same effect.

[highplainsdem]

55. No, it isn't. Writing a story, or writing an article, was never the same thing

as typing something another person wrote or dictated.

Not. Even. Close.

[JCMach1]

63. The reference is to an outdated concept for writing busy work...

That's what ChatGPT is... Just at a different level.

Great for writing boring real estate ads. Crap at writing almost anything meaningful/creative.

It= tool for busywork.

[JCMach1]

64. It's quite good at doing mundane tasks though...

For example, I asked it to write a paragraph about why cats are such a big thing on the internet and it came up with this:

Cats are one of the most popular domestic pets around the world and have been revered for their cute and quirky nature for centuries. On the internet, they have become a sensation due to their adorable antics, expressive faces, and entertaining behavior. Their appeal is widespread and cuts across cultures and ages, making them one of the most relatable animals online. Furthermore, the internet has made it easier for people to share photos and videos of their cats, which has led to the proliferation of cat-themed content. Social media platforms and websites have also capitalized on this trend, with dedicated pages and channels that showcase cats, making them easily accessible to millions of people worldwide. The combination of their popularity as pets and their entertaining online persona has made cats a dominant force on the internet.





ChatGPT Feb 13 Version. Free Research Preview. Our goal is to make AI systems more natural and safe to interact with. Your

Pretty much spot on.

[pwb]

52. What will all these businesses that use it do if we hang up on them?

And they lose business? We don't have to like this right?

[JCMach1]

53. You will never know. ChatGPT is a text tool. Some have also experimented

Using it for coding.

Customer service bots are already there AND you will have no choice with those. You hang up...? No service for you.

[Prairie_Seagull]

57. Some CT scans, MRI scans,and (others one would think) are already being read by people overseas

How long before they are being read by AI. Would anyone know?

How long before some healthcare pros are phased out?

[Meowmee]

58. I am not sure if this is the same thing as chat gpt

I told an essay generating / writing site to write a comparison of two art works, not specific works, but two works each from two parts of a style/ movement. It wrote a well written 2 paragraph or so essay. The first work was correct in info etc. The second work was not. It was wrong, and the description was wrong also. So how is that going to replace humans for this?

I am pretty sure none of my students are using this type of thing because the grammar was too good compared to theirs, and so was the writing structure.

[JCMach1]

65. I asked ChatGPT just now to write a paragraph comparing and contrasting

Rossetti and Whistler... Here's what I got:

In a paragraph compare and contrast Rossetti and whistler

Dante Gabriel Rossetti and James Abbott McNeill Whistler were two influential artists of the 19th century who shared a fascination with beauty and aesthetics. However, their approaches and techniques differed in several ways. Rossetti was a founding member of the Pre-Raphaelite Brotherhood and was known for his romantic, idealized portrayals of women. His paintings featured vivid, jewel-like colors and intricate patterns, often inspired by medieval art and literature. In contrast, Whistler was more interested in exploring the subtleties of light and color, using muted, tonal palettes to create atmospheric, impressionistic scenes. He was a master of Japanese-style composition and his work often featured bold, graphic shapes and flattened perspectives. Additionally, while Rossetti's work often conveyed a sense of narrative and symbolism, Whistler's paintings were more concerned with capturing the momentary sensations of a scene. Ultimately, both artists were significant figures in the development of modern art, but their approaches and techniques differed significantly.

If I was teaching a class on 19th century art, lit, and aesthetics this response would be about a B+ to A- ...

It will only get better. That's the part that freaks out people.

[Meowmee]

68. my result would have gotten a c or maybe even an f

for possible cheating because I know most of my students would never be able to write anything similar and the ones who can have a style. Also it is very generalized and kind of rote. It does not sound to me like a real person wrote it. I always look for personal things in what people write and a style. As I said my result was wrong on one of the works. I gave it more specific parameters than just a general comparison of two artists, and it wasn't able to do that correctly. Again, I am not sure if what I used was chatg etc. On some assignments they have to do a very detailed analysis, and compare to other works, bio, and much more. I doubt any chatg or other format would be able to do it at all. I also have them watch videos and answer specific questions about the videos, while there are ways to cheat on this there is no way chat gpt can do that. We also do narrated or in person presentations, no way chat gpt can do them either.

[JCMach1]

74. All good points... In teaching, the age of rote busywork is done.

If you garbage in, you will get garbage out.

[Celerity]

66. ChatGPT and its ilk are going to be weaponised AF by the RW. Book it. nt

Last edited Mon Feb 27, 2023, 08:59 AM - Edit history (1)

[New Breed Leader]

70. EXACTLY.

all this "if used correctly" bullshit. Spoiler alerts, dum dums: IT'S NOT GOING TO BE USED CORRECTLY.

This is Oppenheimer's bomb.

[Buns_of_Fire]

67. Some companies would do well to replace their CEOs, execs,

and middle-managers with chatbots.

[Johnny2X2X]

72. Productivity

Can you get ChatGPT to spit out nonsense and useless paragraphs? Yes, there are plenty of ways to get it to sound stupid. But If you're aware of how to get it to compile good information and are a willing editor, it can greatly increase your productivity.

AI isn't going to replace your job, a person using AI is going to replace 10 of your jobs.

[ITAL]

73. It's a fun tool to play around with

But it's gonna be awhile until we figure out what it is and isn't good at.

I am a bit of a writer and back in the day I worked in a literary management agency. What is comes up for in regards to creative writing isn't half bad - believe me I read a lot of stuff worse than what this program came up with. It's not particularly good. I do think it can help spur the creative process though. Just playing around with it on a hypothetical screenplay set 150 years a go I gave it all sorts of suggestions on story ideas and the way it spit things back at me so fast gave me a better feel for how I could tackle the subject if I ever did it for real. But it's nowhere near writing a book or script or screenplay yet.