Have you tried ChatGPT?

Technology Technology: Chat / Discussions
, ,
13

And then, just to be topical, I asked it to write me a poem about the Coronation in the style of Oscar Wilde and it produced this doggerel :rofl:

Oscar Wilde would be turning in his grave!

With crown upon his brow and scepter in his hand,
King Charles III sets foot upon the land.
The people gather, their hearts filled with awe,
As the land prepares for a new monarch’s law.

The grandeur and pomp that’s ever so fine,
Unfurls like silken sheets before the divine,
In elegant robes, King Charles III stands tall,
The cheers of the people, echoing with his call.

The trumpets sound, the music fills the air,
The people listen, the choir offering a prayer,
For the land, for the king, for the new beginning,
Their faces uplifted to the heavens, grinning.

With every step he takes, the king shows his might,
Graceful in his movements, a vision of regal sight,
Ever the lord and so just, the goodness in his soul,
Reaching the hearts of all, the people both young and old.

The coronation, a sign of hope and love,
A testament to the power and strength of the dove,
Embracing with a triumphant smile,
The future for which he had long toiled and labored a while.

To witness such beauty, the hearts of all rejoice,
As the new king takes his rightful place, the people’s choice,
May his reign be peaceful, a future bright and fair,
As King Charles III ascends to the throne with regal flare.

14

Last attempt, bored with it now, it just churns out recycled garbled stuff patched together

I asked it to write me a poem about the Coronation in the style of Pam Ayres!

Gather round folks, and listen to me,
For a jolly good tale I’ll weave with wiz’e and glee,
It’s about King Charles - he’s now on the throne,
And we’re all here to cheer and make him known.

There was grandeur and pomp, all the bell and whistles,
With the trumpets blaring and all the officials,
King Charles entered, so tall and fair,
And all the country folk, they stopped and stared.

The choir sang a tune that was quite divine,
And the guests present, they raised a glass of wine,
For a jolly good show it was, you’ll agree,
To welcome the new king in all his majesty.

The scepter he held, and a crown on his head,
The country folk cheered, they whistled and said,
‘Long may he reign and rule with grace,
And bring us peace to this great place.’

For the country they love, their king they adore,
And together they stood, cheering all the more,
With a glint in his eye, and a smile so grand,
King Charles III took the throne, the people’s brand.

So let’s raise our cups and toast to this king,
And the joy that coronations and grandeur bring,
The people united, one big happy family,
As King Charles III leads the nation with such majesty.

15

Oh my word!
:astonished:
Is this what the human race has come to…?

16

Perhaps, but some people are getting concerned.

One of the demands of the Writers Guild in the writer’s strike is limiting use of AI in script writing.

1 Like
17

I don’t think they need be. It has no soul and no inspiration and, like AI paintings, enhanced photos etc us human beings can instinctively sense it

And instinctively, it revolts us, The Uncanny Valley effect

1 Like
18

Can we? The judging panel of this Sony photography contest didn’t sense it.

1 Like
19

When citing sources of information, as is standard in academic work, it simply makes them up.

A fake academic reference from ChatGPT: Mousavi, S. N., Abbasian, S., Khosravi, N., & Asemi, Z. (2022). Timing of nutrient intake and exercise: A systematic review of current evidence and recommendations for future studies. [Note: This academic paper does not exist] Trends in Food Science & Technology, 118, 47-56. [Note: Page numbers are invented]|1024x406.4

“They look perfect - they’ve got the right names of authors, they’ve got the right names of journals, the titles all sound very sensible - they just don’t exist,” James says.

“If you’re not aware of how large language models work, you would be very easily fooled into thinking that these are genuine references.”

:thinking:

Prof Verena Rieser, a computer scientist at Heriot-Watt University who has been working in the AI field for two decades, says chatbots are still in the early stages of development and “can be used to generate misinformation at [a] scale which is obviously very concerning” when it comes to education.

Previous models of ChatGPT were not released because they were deemed “too dangerous”, she says.

“I would expect that we’ll soon see different flavours of ChatGPT by different companies out there and hopefully also safer models which actually mitigate for the possible dangers,” Verena says. “At this moment we don’t really know how to stop the models outputting information which is wrong or toxic or hateful - and that’s a big problem.”

:scream_cat:

2 Likes
20

Much the same could be said for Google. When you put in a search term, it picks up matches even if those are tongue in cheek or biased sources.

Critical analysis is still the cornerstone to research.

If students use ChatGPT with fake references, teachers will be able to spot them faster with a quick search.

2 Likes
21

Will teachers have the time to sift through all the references?

Probably not … they will have to utilise AI to speed up the process.

:wink:

3 Likes
22

Don’t encourage it. :icon_wink: :computer:

23

Thinking. . . so old school. Maybe it will become an anachronism.

24

Could this be an attempt by the establishment to make anything you read on the internet void ?
There are so many people attacking the establishment with theories that could or could not be true. What if all the facts gathered from say Wiki, are neutralised and cancelled out because nobody can ascertain what is true and what is not?..

25

hasn’t that already happened? I’m not sure what happens these days as wiki took some control about ten years ago, but in the old days wiki was a battleground between geeks in basements, who would edit and sub edit entries. These days you have to wonder whether they are using AI to check the script. We will end up with a “the machine stops” scenario.

2 Likes
26

Apparently, Wiki has been using AI since at least 2018:

The Wikipedia community, the free encyclopedia that is built from a model of openly editable content, is notorious for its toxicity. The issue was so bad that the number of active contributors or editors—those that made one edit per month—had fallen by 40 percent during an eight-year period. Even though there’s not one solution to combat this issue, Wikimedia Foundation, the nonprofit that supports Wikipedia, decided to use artificial intelligence to learn more about the problem and consider ways to combat it, including:

  • Collaboration between Wikimedia Foundation and Jigsaw (formerly known as Google Ideas) to Stop Abusive Comments

  • Objective Revision Evaluation Service (ORES System)

  • AI to Write Wikipedia Articles (then underway)

Wiki says:

2 Likes
27

I guess this forum site does the same. There will come a time when you won’t be able to tell whether you are chatting on line with a real human being or a bot.

1 Like
28

I don’t think that will happen on this site but it probably will on Azz’s next creation … :grinning:

29

I thought I would try ChatGPT, just to see what all the fuss is about and I had a number of things I wanted to try, just to see what happens.

However, I had a serious problem with it. I created an account, whilst insisting on giving my phone number, I was then directed to a page where I saw a subscription fee of $7.99 to use it. For some inexplicable reason, my mouse accidentally went to the top of the page and clicked the little X where I quit the site.

So I’m now wondering if this all-powerful AI ChatGPT knows what I mean by “Sticking it where the sun doesn’t shine”.

30

I downloaded the free version to both my phone and my notebook completely free of charge and without any hassle. It’s the one whose database stops in 2019 if I’m not mistaken.

31

Extracts:

It’s a chatbot that responds to almost any prompt, be it a question or command, in convincingly legible prose. GPT stands for Generative Pre-trained Transformer, which means it’s a tool that can generate responses based on what it’s already learned. It is a paid-for tool but there is a free version that you can use if the service isn’t too busy.

The free version of ChatGPT is based on data collection that finished in early 2022, so it does not ‘know’ anything about the world after that time. There’s also a premium version called ChatGPT Plus (which costs $20 a month) that’s furnished with more up-to-date information from GPT-4.

Can I trust the information I get from ChatGPT?

In short, no. ChatGPT can be used in a way that can help get you started and find information that you weren’t aware of. The main thing here is to not use a chatbot as your primary source for information, but instead take the answers it gives you and pursue them until you have found the real facts.

It’s important to understand how ChatGPT comes up with its answers. ChatGPT is very good at placing one word after another. As such, it does not ‘know’ anything at all; all it can do is put words one after another that make sense.

It’s often accurate but equally it can write utter nonsense. The tone ChatGPT uses does not leave any room for doubt, even if it’s talking rubbish and stating facts that are patently false.

If you attempt to probe into how it knows what it’s told you – for instance by asking for a list of citations – it will simply produce a list of things that look like citations but may not actually be real references at all.

Anyone can use ChatGPT for themselves - visit the website, sign up and start experimenting. ChatGPT is a language model, so it can’t generate art or images like some AI engines. However, it can in theory process images and make recommendations. For example, it could scan an image of what’s left in your fridge and then recommend a recipe for dinner.

At your own risk … :wink:

1 Like
32

UK schools have been left confused by the fast rate of change in artificial intelligence (AI) and its impact on education, head teachers are warning.

In a letter to the Times, educators from the state and private sector say developments are “bewildering”. They are launching a body of experts to advise schools on which areas are “beneficial, and which are damaging”.

The technology is moving “far too quickly” for government alone to give adequate advice to schools, they say. AI is the “greatest threat but also potentially the greatest benefit to our students, staff and schools”, the teachers, led by Sir Anthony Seldon, the headteacher of Epsom College, say in the letter.

The group has also questioned the role of digital companies behind AI. “We have no confidence that the large digital companies will be capable of regulating themselves in the interests of students, staff and schools” their letter reads.

Concerns have quickly grown in recent months over AI with the prominence of the ChatGPT bot which has passed exams.

Prime Minister Rishi Sunak recently said that regulation had to evolve at the same time rapid changes are made in AI. He said “guardrails” should be put in place to maximise the benefits of AI while minimising the risks to society.

The group of educators said it was pleased the government was “grasping the nettle” on the issue but felt the need to set up its own body composed of leading teachers “guided by a panel of independent digital and AI experts”.

As with all things technological, kids, who are early users, understand it better then their elders, who are usually latecomers.

1 Like