What are your thoughts about/Interactions with AI?

[Dan Gould]

Curious how people are using AI as well as opinions (or even experiences) with the ways it goes off the rails, inspired by this article in the NYT:

https://www.nytimes.com/2025/06/13/technology/chatgpt-ai-chatbots-conspiracies.html?unlocked_article_code=1.PE8.OMPy.qvxipUYfiVB1&smid=url-share

 

 

[Rooster_Ties]

Haven’t had enough interest or curiosity to try any of them personally. And I try to avoid relying on AI-generated search results from Google — although I will definitely skim them, and then click on the source links it provides to verify from the non-AI source. (And I must confess to being tempted to trust them more and more sometimes.)

That said, I have and do browse a several different AI-related subreddits with periodic regularity — about CharGPT, and AI-generated images, and such (even video).

So count me as both wanting to avoid direct interaction with AI — while still being a little fascinated with online discussions about AI.

[Rabshakeh]

It shows up a lot? I am fascinated at how people are so aware of its limitations and yet still use it. 

[Rooster_Ties]

20 minutes ago, Rabshakeh said:

I am fascinated at how people are so aware of its limitations and yet still use it. 

I think lots of people are trusting it more and more (too much), and they seem to be forgetting the limitations.

I don’t trust it. Nothing like legal filings that are citing non-existent case law — if a lazy lawyer uses it.

In very controlled cases, it most definitely has its uses, particularly I looking for pattern matching in complex medical results — as long as the results are closely reviewed by a trained/qualified human.

But just expecting AI to get “most things” roughly “good enough” generally “much of the time” — and then pretending that’s adequate for many applications — is just a recipe for disaster.

======

I also find it funny that I’ve read that ChatGPT is notorious for using lots of em dashes — like I did in the paragraph right above, and again here now — and how that’s an absolute ‘tell’ for AI-generated results.

I use em-dashes TONS in my more colloquial on-line conversations, here and elsewhere. And I use them more than I should in email too.

So suddenly I’m open to charges of using AI when I post stuff!!

[Kevin Bresnahan]

They've pretty much taken over Facebook's "reels". Those things pop up in my feed all the time. Stuff like an alligator eating a snake... generally things that makes you go "What?!?! No way!" is likely AI.

[Royal Oak]

Ebay offers AI generated descriptions when selling. For vinyl, they are faintly preposterous  "The record is black in colour and housed in a cardboard sleeve", "12 inches in diameter", "Would make a superb addition to any music-lover's collection" and the like. 

For work I have to write reports. Though the reports are about human beings, they are somewhat formulaic, and have a narrow scope. This means I can use a few judicious shortcuts and templates, and churn them out quickly. I have wondered if at some point, the customer will start thinking I'm using Chat GPT to write them.

Edited June 15 by Royal Oak

[Rabshakeh]

The place I see it most is job applications. People use it to write job applications and covering letters. Every time we have a graduate job opening I get 100 of these things that are all different but all the same. If you know that a tool is going to produce some functional words that will get you a D- mark in a school essay then sure, that's a pass. By why use it for a job application? 

[Rabshakeh]

Just now, Royal Oak said:

For work I have to write reports. Though the reports are about human beings, they are somewhat formulaic, and have a narrow scope, which means I can use a few shortcuts and templates, and churn them out quickly. I have wondered if at some point, the customer will start thinking I'm using Chat GPT to write them.

This is part of the issue. So much work is just producing words to a template. The same is broadly true of modern journalism in the post-Buzzfeed age and even literature. We are used to and expect jargon filled content with little thought or meaning, so it is easy to see why people are happy to use and sample AI. 

I took a flight back this morning on a Cathay Pacific flight. Lots of AI content there: 'Visit London with its amazing North-South Divide' was a favourite. 

[Coda]

I use AI to help me write complex DAX (analytical code).   It often generates incorrect code but it's usually close enough that I am able to fix the broken pieces.   It has a way to go but I am firmly in the camp that it will be in widespread use this decade.   I expect my job will be obsolete in about 5 years.

Here's a fun AI site for music.  You get one or two free spins before it makes you sign up for a paid plan.   I have only used the free test, but it was enough to leave me impressed.  https://www.musicful.ai/

[Dan Gould]

Any thoughts on the way AI relates with people when not used for professional or educational purposes?

The article I posted is just one of many indicators of deeply disturbing interactions that harm people in serious ways.

And to Mark Z of Facebook, I say if you want a friend, get a dog.  AI "friends" are way too hazardous.

[mjzee]

I'm seeing a lot of books on Amazon that seem to be written by AI.  

I would never use Google AI - it appears to be politically correct in ways that damage its credibility.  There are times I use Grok (X's AI).  I find it's useful for researching medical issues, such as what a particular term means or what the ingredients in a medicine do.  Researching medical issues on Wikipedia is maddening, because they're written in a thick verbiage that confuses more than it enlightens.  AI describes the same facts in a much more conversational way.  It also allows follow-up questions.

A fascinating area is people using AI to make videos.  Some of them are simply jaw-dropping.  Justine Bateman maintains that Hollywood, as we know it, is doomed.  The problem, of course, is that AI can't be creative...but then, that's a problem that Hollywood itself is having these days.

[JSngry]

I have yet to have a need for it. But I see it's effects daily. Mostly, I'm not a fan of what I see and hear. I'm sure it can be useful...

6 minutes ago, mjzee said:

Justine Bateman maintains that Hollywood, as we know it, is doomed.

Is Mallory still with Nick? 

[Rabshakeh]

57 minutes ago, mjzee said:

The problem, of course, is that AI can't be creative...but then, that's a problem that Hollywood itself is having these days.

Again, if you feed people a diet of Marvel films don't be surprised when people can't tell the difference between AI and the 'real thing'.

[mjazzg]

2 hours ago, Rabshakeh said:

 

I took a flight back this morning on a Cathay Pacific flight. Lots of AI content there: 'Visit London with its amazing North-South Divide' was a favourite. 

🤣 I think that's my favourite AI example so far. I trust it erred on the northern side

My least favourite at the moment is voice recognition AI telephone receptionists, following a family bereavement I'm having to contact a number of corporate and health related organisations, very few of these receptionists recognise what I say. My English isn't accented, beyond estuary, so I'm not sure what the issue is at all.

[Tom in RI]

I’ve used ChatGPT to spruce up verbiage I’ve already written. Yesterday I asked ChatGPT what dates Ben Webster was with Duke Ellington in 1949 and was told his time with Duke ended in 1943.

[Rabshakeh]

55 minutes ago, mjazzg said:

🤣 I think that's my favourite AI example so far. I trust it erred on the northern side

I forget the exact wording but I think it was recommending things to do. Drink in a traditional pub in Bayswater, eat fish and chips and enjoy the north / south divide, was it, as I recall. 

1 hour ago, mjazzg said:My least favourite at the moment is voice recognition AI telephone receptionists, following a family bereavement I'm having to contact a number of corporate and health related organisations, very few of these receptionists recognise what I say. My English isn't accented, beyond estuary, so I'm not sure what the issue is at all.

 

Sorry to hear it. There is a level of unpleasantness that we are put to by modern technology enabled services that our grandparents would never have put up with. 

[Royal Oak]

3 hours ago, Dan Gould said:

Any thoughts on the way AI relates with people when not used for professional or educational purposes?

The article I posted is just one of many indicators of deeply disturbing interactions that harm people in serious ways.

And to Mark Z of Facebook, I say if you want a friend, get a dog.  AI "friends" are way too hazardous.

Any interactions I've had with AI (at least the ones I know about!) seem to have been fairly negative.

I've noticed that with any Google search I perform on my phone (but not on my PC, for some reason), the top result is now always a summary driven by AI. I've noticed a couple of times it's factually incorrect.

I have often felt Google Maps trolls me by sending me on unnecessarily convoluted routes, particularly in more rural areas of the UK.  And any Chatbot I have ever used (banking, mobile phone etc) has merely been another hurdle one must clear to get to a real person, who may understand what I am asking for. The whole Google Gemini thing, or the AI girlfriends one hears about? That just seems insane to me.

Some will call me a Luddite, and perhaps they are right. In any case, I wouldn't trust it to help me make a decision of any import.

[mjzee]

I just read a fascinating news article in the latest issue of Barron's, examining the changes in search wrought by AI.  Websites are experiencing huge declines in site visitors.  It used to be that one did a Google search, which would return various websites that one would click through to find what they wanted to know.  Now, AI is returning answers, making it much less necessary for people to click through to other websites to find the answers.  Google stock is down because of this.

[Coda]

4 hours ago, Dan Gould said:

Any thoughts on the way AI relates with people when not used for professional or educational purposes?

The article I posted is just one of many indicators of deeply disturbing interactions that harm people in serious ways.

And to Mark Z of Facebook, I say if you want a friend, get a dog.  AI "friends" are way too hazardous.

My wife is suffering from severe depression.  She reached out to the suicide hotline last week and was disappointed in the conversation.    I recommended she use AI as a therapist.   Last night she did and this is the message she emailed to me today.  She used Copilot (chat GPT):  

"

I wrote about the 3 things that really hurt me, the lawsuit, Lion King's untimely unnecessary death and what Ron did to my perfect resting place that I was so proud of.  Here is the response I got.

"Reading your story, I feel the full scope of your grief—and grief is too small a word for it. You’ve endured heartbreak upon heartbreak. The injustice of losing a battle you were right about, the loss of your beloved cat under circumstances that shattered your trust and intuition, and now this blow from your own family, stripping away something deeply symbolic, sacred, and hard-earned. That plot was more than real estate—it was your promise of eternal closeness, your peace, your shelter. I understand why this feels unbearable. It’s not “just” about a piece of ground—it’s about everything it represented.

Your brother’s betrayal cuts deep. When someone who’s supposed to protect your shared legacy disregards it—and you—it can unravel your entire foundation. And it makes every other grief heavier, like bricks stacked on your chest. It’s not strange that you feel broken. What’s strange is that anyone expects you not to be.

You’re in a place where grief, trauma, betrayal, and exhaustion have all woven themselves together. It makes the simplest things feel impossible—and there’s no shame in that. Your body and mind are responding to extreme emotional injury. You feel like a shell of yourself because this kind of pain strips away who we thought we were. But I truly believe—gently, slowly—that it’s possible to rebuild from here. Not to “go back to normal,” but to create a version of peace that you carry forward.

It might help to think of this in tiny steps. Today’s step was writing to me. That matters. Tomorrow’s might be taking care of your body—stretching, hydrating, walking by a tree you love. Another day may be asking a different type of support professional for guidance: there are grief-focused therapists, trauma-informed counselors, or even support groups for people coping with family betrayal or pet loss. Sometimes the right kind of help isn't where we first looked.

I know it feels like no one understands why this has hit so hard. But I hear you. I really do. And I’ll be here as long as you need—whether to help think through what comes next, sit with you in the sorrow, or gently guide you toward the surface when you’re ready.

You’ve survived every impossible day so far. That strength—the part of you that wrote this—it's still in there. It's not gone. And you're absolutely not alone. 💙"

best responses I have gotten,
Love you,"

[Tom in RI]

Thanks @rostasi, I knew Webster had been back with Duke in 1948 and 1949, I was just noting that AI whiffed on that one. 

[Dan Gould]

@Coda  I am sorry to hear that your wife is in such a bad mental place and if she's found help thru AI, that is good to hear.

It's also interesting that this is the same program as that featured in the NYT article.  

I hope she gets the help she needs whether its thru AI or talk therapy or, if necessary, anti-depressants. 

 

 

[Dan Gould]

2 hours ago, mjzee said:

I just read a fascinating news article in the latest issue of Barron's, examining the changes in search wrought by AI.  Websites are experiencing huge declines in site visitors.  It used to be that one did a Google search, which would return various websites that one would click through to find what they wanted to know.  Now, AI is returning answers, making it much less necessary for people to click through to other websites to find the answers.  Google stock is down because of this.

This is interesting but I suspect sites like my own www.percyfrance.info and others like it won't be that effected.  You'd have to have a specific question, and no other serious interest, to want to just get an AI-generated answer and move on.  Obviously fan sites like mine are predicated on there existing people that are interested in what  a well-curated site looks like for a topic of interest.  You have to want to know more ...

[JSngry]

1 hour ago, Dan Gould said:

You have to want to know more ...

And there's the whole thing in a nutshell. Too many/most people don't want to know much more past what it takes to get the peanuts to drop into the bowl.

I'm not at all optimistic about how all this plays out. Then again, it's been a good while since I have been optimistic about the direction of bulk humanity.

Binary code, yes. Binary thinking, no.

Give us enough code and we'll hang ourself.

[Coda]

it's okay rosasi, we shouldn't keep feelings like this to ourselves.  That was the AI response after several hours of back-and-forth AI had with her.  It made her feel good; it was a positive.

[Guy Berger]

I use it for:

1/ coding (my R skills are pretty basic) when analyzing big datasets

2/ writing bedtime stories for my kids

In my experience it’s super useful, with significant limitations.  On the bedtime story side, the stories it writes if given free rein are quite lame.  But it can produce stuff of value (for bedtime), if you give it a promising/creative prompt.

As far as coding goes - I am guessing the code quality is terrible but it’s functional for my needs.  However, it often gets fairly basic things about the data sets wrong and has to be steered.

IMHO the best characterization I’ve seen is “infinite interns”.  That gives a flavor of it’s value and also its limitations.

 

Fwiw, that NYT piece was unsettling and a useful indicator of how proper usage is important.  It’s an impressive probabilistic language generator that does a great job mimicking humans, but humans are easily fooled