So, how worried should we be about AI?

Whitewavemark2

Did you see the art photo entry done as it turns out by a machine? I think that it can also produce music in the style of a great composer.

I may be missing a point, but to me this has no value or point, as what we value and admire is the human creativity contained in the piece. I fail entirely to see the value of machine produced stuff.

You’re totally right.
AI are now going to write books as well, what it does is takes ideas, tropes and writing from existing books and stitches them together.Certain authors are already doing this and there will soon be no original thought, writing, music, poetry at all if it isn’t reined in.It’s new but at the same time has been known about for a while so there should have been laws made and potential bans ready to use.
It will put authors and musicians, journalists and many other professions out of work and the arts will be worse for it.
As other say too, we won’t know if articles or photos or anything is real either.It’s terrifying.

I don't want to lower the tone, but I had a vivid dream about this in which robots were completely human-like.
Even then there were class differences, - the ordinary masses were known as FM - Factory Made. But the elite were known as HM - Hand Made. It was a scary dream!

stewaris

growstuff
It appears google appears to think it can. See here futurism.com/the-byte/google-ai-bengali

Sorry, I can't add the hyperlink.

I'm with growstuff on this.

My google knows it can't.

analyticsindiamag.com/did-google-bard-really-learn-bengali-on-its-own/

Message deleted by Gransnet. Here's a link to our Talk guidelines.

growstuff
It appears google appears to think it can. See here futurism.com/the-byte/google-ai-bengali

Sorry, I can't add the hyperlink.

stewaris

MeryStreep The problem is we are no longer talking about AI in it's predictive form. We are talking about machines that can teach themselves languages. If they can do that now what are they capable of in the future.

They can't teach themselves languages. They can operate in different languages, which isn't the same thing as learning a language.

Translating apps have become increasingly sophisticated over relatively few years and are commonplace.

VERY worried!

MeryStreep The problem is we are no longer talking about AI in it's predictive form. We are talking about machines that can teach themselves languages. If they can do that now what are they capable of in the future.

It is not AI that I am worried about it is the use that AI will be put to that worries me.

The Economist has a headline today

“How to worry wisely about AI”

So that worries me!

It seems to have a very comprehensive article - I might just join up for a month to read it later.

The above technique is already being used for political trolling on social media sites, which is how there are sometimes so many allegedly supportive messages about a particular politician or initiative. They're not written by real people but "bots" using AI.

stewaris

MerylStreep this device is the only access to the internet I have and I access all sites incognito, check all cookies and clear them out regularly. I also have an ad blocker and subscribe to internet protection. As well as that, my nephew works in IT and runs scans on all the family's laptops etc on a regular basis. This is one of the very few sites I access on a regular basis and the only one I occasionally post on.

MerylStreep is correct. AI isn't a virus. There are a number of legitimate AI sites which no virus checker or adblocker would identify.

One of them is called chatgpt. I could, for example, log on to it and instruct it to write a post in the style of stewaris on Gransnet. It would analyse any posts you've written in the past and produce a Gransnet-style message. If I wanted to be malicious I could sign up with a username almost identical to yours and post my message and nobody, except you, would realise that you hadn't posted it.

My worry is the misuse of AI. My friend had a simulated phone call. She believed it was from her teenage GS asking her to take money to the police station to bail him out. (sorry I can't remember the exact details) She had the good sense to phone her son who said GS was at home. My friend was astonished that the simulated voice sounded EXACTLY like her grandson's voice. These scammers are damned clever unfortunately.

Stewaris
What do think predictive text is?
AI in its simplest basic form. It’s not a virus that can be detected by a scan.
It’s built into the software.

Due to all the above comments, I just can't help thinking of technophobia - and Luddites!

MerylStreep this device is the only access to the internet I have and I access all sites incognito, check all cookies and clear them out regularly. I also have an ad blocker and subscribe to internet protection. As well as that, my nephew works in IT and runs scans on all the family's laptops etc on a regular basis. This is one of the very few sites I access on a regular basis and the only one I occasionally post on.

Yes, I was thinking about that programme, very scary

to answer the title question.. Yes! glad I should not be long for this world

This week the Danish school-teachers' unions have asked the Ministry of Education to provide guidelines for when and how much pupils may use chatbots and other forms of artificial intelligence when doing homework, or preparing term projects.

The universities, too, are looking into the legal ramifications of it, as obviously a paper complied by a chatbot is not a student or faculty member's own work, which has always been a prerequisite of academic work.

The ministry of Defence is likewise worried - with good reason. Since the start of the war in Ukraine and seeing the amount of propaganda being generated or at least sent round on social media, I have felt and said that the world can be thankful that that master of propaganda Doktor Joseph Goebels did not have the Internet to play with!

Stewaris
i would never allow anything like that in my home
Then I suggest you get rid of every device that uses the internet, including the one you’re using to post on here.

dataconomy.com/blog/2022/05/09/artificial-intelligence-in-everyday-life/?utm_content=cmp-true

Frankly, I think the whole thing quite scary. I read in the paper they (the great and the good) are saying they need to get together and workout a set of guidelines globally to ensure it won't be misused - it will be. Shades of Terminator:The Rise of the Machines. I would never allow anything like that in my home. I don't have Alexa either.

I have been conned by shallow fakes - so no chance with deep ones!

This links up with a previous thread on deep fakes which are equally terrifying:
www.gransnet.com/forums/scams_and_fraud/1313313-Dont-be-conned-by-a-deepfake

MaizieD

^Ultimately, the future impact of AI will depend on how we choose to develop and deploy this technology in our society.^

That's not a very balanced answer, is it?

You can bet your life that there are people even now who are using, or intend to use, AI for rather less than noble purposes.. I think that AI should be capable of acknowledging that. Perhaps it is but doesn't want us to know...

Students use AI to write very plausible essays. Teachers and lecturers know about it and are trying to change their assessment methods.

No - what I really meant was it was exactly the answer I would expect from AI. - so good in that sense😊