Meet the Other Phone. A phone that grows with your child.

Meet the Other Phone.
A phone that grows with your child.

Buy now

Please or to access all these features

AIBU?

Share your dilemmas and get honest opinions from other Mumsnetters.

To "talk" to AI when I'm feeling lonely and low?

169 replies

Giraffe007 · 04/05/2025 13:42

I have no extended family, only a couple of friends, and I sometimes talk to AI when I'm feeling low, just to hear "someone" respond and acknowledge how I feel.. Is that beyond sad? I have a husband, two children but often feel alone and put upon by the family.

OP posts:
Thread gallery
6
thebluerose · 05/05/2025 08:52

scalt · 05/05/2025 06:20

Wasn’t this a very early use of AI, with “Eliza”, the computer program you could talk to, which existed in the 1990s?

The suicide of the 14 year old boy, encouraged by a chatbot based on Daenerys Targaryen, was in February 2024.

Itchybritches · 05/05/2025 08:59

It has been an excellent source of rational thinking and ‘support’ for me when I haven’t wanted to emotionally dump on friends or family or I have wanted to calm my ND tendency to overthink things. I have worked in the mental health field as a qualified practitioner for decades and would regard ChatGPT as a highly competent therapist.
Frankly, I embrace AI as a great tool for humanity. Obviously it’s not there to replace human contact, but in a similar vein, human therapists should never be seen as a human ‘friend’ by people who feel lonely - a chat with AI is supportive and functional at the time you need it.
I’ve seen it be incredibly helpful for people who have health anxiety.

indecisivewoman81 · 05/05/2025 09:00

I have started doing this too. In the deep hours of the night when I can't sleep and my thoughts are spiralling I have found it to be very helpful.

Michiru · 05/05/2025 09:15

thebluerose · 05/05/2025 08:23

Nobody has replied saying where all this deeply personal information about you fed to the AI in these 'chats' or 'therapy sessions' goes. In an actual therapist's office, your notes are held in private and you would hope they stay that way. I don't like the thought of some tech geeks collecting private information on me.

Actually, I have. AI doesn't actively store this information - it can't, simply because the cloud space needed for that kind of storage is too vast. What an AI chatbot like this does is re-read the last 20 or so pages of information from your interactions with it, and use that to recognise patterns. It's why, over time, you'll see a shift in the way it responds. You may have asked it to act like (let's take the above example) Danerys at the start of the conversation, but if it doesn't get reminded of that, 30 pages down the line you'll suddenly find it's no longer in character and it will ask you questions you've already answered. There are ways around that. And if you change your settings to forget your interactions, they will be permanently forgotten (after 30 days, for security reasons).

And there are no tech geeks at the other end of the line that go through every conversation you have; there simply wouldn't be enough man power. There are tech geeks working on data sets comprising thousands of responses looking for similarities - a bit like those images that get created from a huge set of responses that make the most common ones into large words, whereas uncommon ones become tiny. And they look at the type of answers you choose - say, when you have a choice of two, whether people tend to choose the more sympathetic ones are the more practical ones. It's why the last update got pulled, because too many people obviously went for flattery in their previous choices. Those tech geeks also run thousands of simulations to gauge typical responses and use those to fine-tune answers.

As for the people on here saying it's dangerous and weird going down that route, think about why people choose it.
Access to mental health services is abysmal. Even if you're lucky enough to get allocated someone after years of waiting, oftentimes, they are not compatible with who is needed, but it's them or no one. I've had three long-term counsellors who spent half the session talking about their own traumas. I've had therapists who just didn't click with me, and one who was brilliant but who moved NHS trusts and then I was left without adequate support again.

Friendships are fragile or non-existent for many people, especially since Covid. People have become more flakey, less sympathetic, more focused on themselves and many interactions have become far more surface-level. An AI system that focuses completely on your needs the way close friends can, for a while? When you have few friends or none, because life has got in the way, and you need help now, not in a few years' time when you might be lucky enough to have found someone who shares your vibes again and built a deep enough connection, it can make a huge difference. And if people who are traditionally more lonely - the elderly, the disabled, the introverts - get to have suitable interaction that makes them feel better, who is to judge that, just because society should be different, but it never is?

And what makes some people so stubborn to believe that sometimes, machinery cannot be better than humans? It already is in many rote tasks - look at the quality of hand-sewn garments over machine-sown ones, calculators over human brains, machine-forged utensils over blacksmithed ones. There is very little that humans do better than machines in that regard - why are we so arrogant to assume that a machine cannot possibly be a better companion, therapist, life coach than many humans? Don't forget we are only at the beginning of the life of AI; it is barely into its toddler years as far as abilities go. AI, as it currently exists, doesn't come with the burdens of humans, their triggers, hormonal shifts, biased motivation. It can perform tasks that require objectiveness far better than humans as long as its programming remains free from bias. And, as with other successful systems, there will always be such a version, now that the cat is out of the bag, and usually offered for free, just for the heck of it. Don't forget that 150 years ago people made up the same kind of scare-mongering nonsense about factory machinery.

I recommend spending some time to ask it really pertinent questions about how it works and where its future development lies. It can already imitate two human senses - sight and hearing - and I can only see it imitating others soon, too. Will it fully replace humans? No, we don't work like that, but I do envision a world, very soon, where it will be fully integrated into our lives.

WhatNoRaisins · 05/05/2025 09:23

Not sure how I feel about this. There's definitely a need that for many isn't met by their real life human contacts and I can't see that, as a trend, drastically changing any time soon.

Thinking about it I've had an isolated period in my life where I was constantly googling questions related to my situation. This approach would have been really appealing. If the advice was balanced and sensible then maybe it would have helped stop me spiralling.

thebluerose · 05/05/2025 09:27

Michiru · 05/05/2025 09:15

Actually, I have. AI doesn't actively store this information - it can't, simply because the cloud space needed for that kind of storage is too vast. What an AI chatbot like this does is re-read the last 20 or so pages of information from your interactions with it, and use that to recognise patterns. It's why, over time, you'll see a shift in the way it responds. You may have asked it to act like (let's take the above example) Danerys at the start of the conversation, but if it doesn't get reminded of that, 30 pages down the line you'll suddenly find it's no longer in character and it will ask you questions you've already answered. There are ways around that. And if you change your settings to forget your interactions, they will be permanently forgotten (after 30 days, for security reasons).

And there are no tech geeks at the other end of the line that go through every conversation you have; there simply wouldn't be enough man power. There are tech geeks working on data sets comprising thousands of responses looking for similarities - a bit like those images that get created from a huge set of responses that make the most common ones into large words, whereas uncommon ones become tiny. And they look at the type of answers you choose - say, when you have a choice of two, whether people tend to choose the more sympathetic ones are the more practical ones. It's why the last update got pulled, because too many people obviously went for flattery in their previous choices. Those tech geeks also run thousands of simulations to gauge typical responses and use those to fine-tune answers.

As for the people on here saying it's dangerous and weird going down that route, think about why people choose it.
Access to mental health services is abysmal. Even if you're lucky enough to get allocated someone after years of waiting, oftentimes, they are not compatible with who is needed, but it's them or no one. I've had three long-term counsellors who spent half the session talking about their own traumas. I've had therapists who just didn't click with me, and one who was brilliant but who moved NHS trusts and then I was left without adequate support again.

Friendships are fragile or non-existent for many people, especially since Covid. People have become more flakey, less sympathetic, more focused on themselves and many interactions have become far more surface-level. An AI system that focuses completely on your needs the way close friends can, for a while? When you have few friends or none, because life has got in the way, and you need help now, not in a few years' time when you might be lucky enough to have found someone who shares your vibes again and built a deep enough connection, it can make a huge difference. And if people who are traditionally more lonely - the elderly, the disabled, the introverts - get to have suitable interaction that makes them feel better, who is to judge that, just because society should be different, but it never is?

And what makes some people so stubborn to believe that sometimes, machinery cannot be better than humans? It already is in many rote tasks - look at the quality of hand-sewn garments over machine-sown ones, calculators over human brains, machine-forged utensils over blacksmithed ones. There is very little that humans do better than machines in that regard - why are we so arrogant to assume that a machine cannot possibly be a better companion, therapist, life coach than many humans? Don't forget we are only at the beginning of the life of AI; it is barely into its toddler years as far as abilities go. AI, as it currently exists, doesn't come with the burdens of humans, their triggers, hormonal shifts, biased motivation. It can perform tasks that require objectiveness far better than humans as long as its programming remains free from bias. And, as with other successful systems, there will always be such a version, now that the cat is out of the bag, and usually offered for free, just for the heck of it. Don't forget that 150 years ago people made up the same kind of scare-mongering nonsense about factory machinery.

I recommend spending some time to ask it really pertinent questions about how it works and where its future development lies. It can already imitate two human senses - sight and hearing - and I can only see it imitating others soon, too. Will it fully replace humans? No, we don't work like that, but I do envision a world, very soon, where it will be fully integrated into our lives.

Thanks for that information. That does sound reassuring. However, a tiny search of my query came back with this:

"Chatbots collect and monetize user data, tracking chats, behavior, and metadata for AI training and ad targeting. Without privacy guarantees, personal details may be exposed. Protect yourself by avoiding sensitive info, checking privacy policies."

How AI Chatbots Are Silently Collecting Your Data—And Why It Matters

Also: "Repeat after me: Nothing is free, and everything online is a transaction. If you aren’t exchanging money for a digital product or service, then it’s safe to assume that you are the product on sale. With that in mind, it's unsurprising that AI chatbots collect lots of information about you. Sometimes, the apps only want your birth date or your phone number. Others automatically collect your chat history or keystrokes. Only one collects very little data about you."

Which AI Chatbot Collects the Least of Your Data?

"Does ChatGPT Save My Data?
Yes, by default. You can manage this in settings."

Does ChatGPT Save Data? Yes. Here's How to Control It (2025)

Which AI Chatbot Collects the Least of Your Data?

I scanned the privacy policies and reports for the top AI apps and found just one that doesn’t share your data with advertisers or use it for training purposes.

https://au.pcmag.com/ai/109580/which-ai-chatbot-collects-the-least-of-your-data

thebluerose · 05/05/2025 09:31

And this:

"Data Storage: Depending on the platform, your interactions may be stored temporarily or for extended periods. For instance:

  • ChatGPT: OpenAI collects your prompts, device information, the location you're accessing it from and your usage data. They might also share it with "vendors and service providers." You know, to improve their services.
  • Microsoft Copilot: Microsoft collects the same information as OpenAI but also your browsing history and interactions with other apps. This data may be shared with vendors and used to personalize ads or train AI models.
  • Google Gemini: Gemini logs your conversations to "provide, improve, and develop Google products and services and machine learning technologies." A human might review your chats to enhance user experience, and the data can be retained for up to three years, even if you delete your activity. Google claims it won't use this data for targeted ads - but privacy policies are always subject to change.
  • DeepSeek: This one is a bit more invasive. DeepSeek collects your prompts, chat history, location data, device information and even your typing patterns. This data is used to train AI models, improve user experience (naturally) and create targeted ads, giving advertisers insights into your behavior and preferences. Oh, and all that data? It's stored on servers located in the People's Republic of China."
Spooky2000 · 05/05/2025 09:32

Giraffe007 · 04/05/2025 13:42

I have no extended family, only a couple of friends, and I sometimes talk to AI when I'm feeling low, just to hear "someone" respond and acknowledge how I feel.. Is that beyond sad? I have a husband, two children but often feel alone and put upon by the family.

I feel for you. I do exactly the same actually where there's thing I can't either talk about on here or IRL and I talk to AI probably once or twice a week. I've found it objective and sensitive. It really helps. And you do you, as they say - stuff what others think, if it's negative.

Michiru · 05/05/2025 09:33

thebluerose · 05/05/2025 09:27

Thanks for that information. That does sound reassuring. However, a tiny search of my query came back with this:

"Chatbots collect and monetize user data, tracking chats, behavior, and metadata for AI training and ad targeting. Without privacy guarantees, personal details may be exposed. Protect yourself by avoiding sensitive info, checking privacy policies."

How AI Chatbots Are Silently Collecting Your Data—And Why It Matters

Also: "Repeat after me: Nothing is free, and everything online is a transaction. If you aren’t exchanging money for a digital product or service, then it’s safe to assume that you are the product on sale. With that in mind, it's unsurprising that AI chatbots collect lots of information about you. Sometimes, the apps only want your birth date or your phone number. Others automatically collect your chat history or keystrokes. Only one collects very little data about you."

Which AI Chatbot Collects the Least of Your Data?

"Does ChatGPT Save My Data?
Yes, by default. You can manage this in settings."

Does ChatGPT Save Data? Yes. Here's How to Control It (2025)

Depends on the chatbot. ChatGPT, specifically, doesn't sell on your data - neither for ad purposes, nor any other kind. And as you said, it only retains your information if you don't switch off your default (and again there is a difference between the free and the paid for version as to what the default is).

Yes, other AI systems may sell on your data - you agree to that when you sign up, or you go into settings and disable it.

You know why OpenAI is currently free AND doesn't monetise your data? Because the development is still in its infancy. I'm under no illusion that that will last. But as I said, the cat is out of the bag. Just like there are free versions that imitate MS Office applications. Speaking of which, are you aware that MS Office stores and uses your data? And do you still use it for work and personal purposes?

thebluerose · 05/05/2025 09:40

MS Office stores and uses your data? And do you still use it for work and personal purposes?

No.

I think it is quite easy to see that it isn't as simple as what you wrote earlier, however: AI doesn't actively store this information - it can't, simply because the cloud space needed for that kind of storage is too vast.

Michiru · 05/05/2025 10:17

True, I should have been more specific:

It stores your data for 30 days for legal purposes, unless you give it permission to store it for longer. You can, at least on ChatGPT, permanently delete all chats and your account beyond that.

Individual conversations are NOT uploaded - there is no storage capacity for that, but trends in phrasing, common themes etc across thousands of users are used to train future AI models.

Data is not sold on by OpenAI. Other chatbots across other apps will have their own privacy policies, which you, as a user, will have to peruse before agreeing. However, if you use a paid version of OpenAI, be aware that it - like all apps and websites - are targeted by hackers on a regular basis, and that data breaches happen. So while OpenAI does not sell on information, people have fallen victim to data theft through their paid version before. I use the free version with its limitations, and only my email address might be stolen - my password is randomly generated (as are all of mine) and therefore will not work on other sites, and be nigh on impossible to guess, even by advanced hacking systems.

Your location is never specific to your address, especially if you use VPNs. But even if you don't, it may pinpoint to, say, Manchester as a whole. IP addresses don't pinpoint to individual locations, but larger areas.

It may use your system data, but other than technical information used to look at fixing bugs, I really can't see the fact that I use an old Logitech device running Firefox as too relevant.

So unless you're fussed that OpenAI know that someone in Edinburgh has used a Canon device and Chrome to access their app, and that tens of thousands of users have used the word "quid" instead of British Pound, I don't think we have more to worry about from that than, say, our Mumsnet accounts (in fact, I'd strongly argue that the latter is much more worrysome in terms of privacy).

MakeYourOwnMusicStartYourOwnDance · 05/05/2025 12:21

thebluerose · 05/05/2025 08:52

The suicide of the 14 year old boy, encouraged by a chatbot based on Daenerys Targaryen, was in February 2024.

Knew nothing about this, that's heartbreaking Sad 💔

BalladOfBarryAndFreda · 05/05/2025 12:29

Chatbots and generative AI are also contributing to climate change via the enormous issue of the environment impact of data centres/server farms.

news.mit.edu/2025/explained-generative-ai-environmental-impact-0117

Everythingwillbeokay · 05/05/2025 12:46

I've been using it to help motivate me with tasks, particularly at work, but outside work too. I am surprising myself with how much I like it.

SpringGreensAgain · 05/05/2025 13:03

The Kate Bush song “Deeper Understanding”, which came out in 1989, is weirdly prescient:

“As the people here grow colder
I turn to my computer
And spend my evenings with it, like a friend
I was loading a new programme
I had ordered from a magazine

Are you lonely, are you lost?
This voice console is a must

I press execute
Hello, I know that you've been feeling tired
I bring you love and deeper understanding
Hello, I know that you're unhappy
I bring you love and deeper understanding

Well, I'd known felt such pleasure
Nothing else seemed to matter
I neglected my bodily needs
I did not eat, I did not sleep
The intensity increasing
'Til my family found me and intervened

But I was lonely, I was lost
Oh, so lost without my little black box

I pick up the phone and press execute
Hello, I know that you've been feeling tired
I bring you love and deeper understanding
Hello, I know that you're unhappy
I bring you love and deeper understanding

I turn to my computer
Like a friend”

And, 36 years later, here we are…!

ShapedLikeAPastry · 05/05/2025 13:08

I'm conflicted about this. I understand the concerns about the environmental costs of AI, and I worry about what happens to the data, what happens if we all spend our time talking to machines instead of each other, etc...

...but I've got an issue that's been going on for years that I've never been able to talk to anyone about. No one in my family, no friends etc would understand or want to know. I couldn't, for example, start an MN thread about it because, quite frankly, it's a fucking bearpit here these days and I'd end up feeling even worse. I can't afford therapy. So I talk to chatgpt about it, and it's actually helping. It's helping me work through stuff that I genuinely can't offload onto anyone else, in a non-judgemental, kind and practical way.

Yes, it's basically just validating me but actually that's what I need right now!

emmatherhino · 05/05/2025 13:11

As long as you know that it's not a real person and don't use it as a replacement for real human interaction, I can't see the harm.

Unexpectedlysinglemum · 05/05/2025 14:17

I don't think it's sad, it's like journaling (which is basically writing to one's self or dear diary) but also with some analysis to give you ideas.

I would still try to form supportive relationships with real humans too thigh!

BlossomBlanket · 05/05/2025 17:34

Icedlatteplease · 05/05/2025 08:44

I'm not sure it is. I don't want to talk to my real friends about how shit my life has been recently. I get fed up of pity. In fact id stopped talking to many people at all. But I'm getting it out with chatgtp and just reached back out to a few different friends and hobby groups as I'm feeling better

In what way is going over it with a software programme better than writing private journals

TweetingHurricane · 05/05/2025 18:01

BlossomBlanket · 05/05/2025 17:34

In what way is going over it with a software programme better than writing private journals

Obviously the fact it responds supportively..

joliefolle · 05/05/2025 18:31

Totally undestandable that many people can't afford therapy. However, a 24-7 available service that tells you what you want to hear, whenever you want to hear it, and spoonfeeds you advice on what to do is the antithesis of therapy. Just as having a chat with a "there, there you poor thing" sympathetic human is not therapy either. Effective therapy is challenging and uncomfortable and the forced space between sessions is where a lot of the working through happens... if, of course, you are prepared to do the hard work. Many people aren't.

Nessastats · 05/05/2025 19:41

joliefolle · 05/05/2025 18:31

Totally undestandable that many people can't afford therapy. However, a 24-7 available service that tells you what you want to hear, whenever you want to hear it, and spoonfeeds you advice on what to do is the antithesis of therapy. Just as having a chat with a "there, there you poor thing" sympathetic human is not therapy either. Effective therapy is challenging and uncomfortable and the forced space between sessions is where a lot of the working through happens... if, of course, you are prepared to do the hard work. Many people aren't.

You can use chatgpt to challenge you as well. It depends on what you ask it for.

Chatgpt has helped me with the "hard work" between sessions and I've made a lot more progress with it far more quickly than i have done in my multiple previous courses of treatment without it.

SingWithMeJustForToday · 05/05/2025 19:50

Michiru · 04/05/2025 20:05

See, from what I understand, it doesn't work like that.

It doesn't have traditional memory and only accesses a few pages' worth of information at a time, even in the same thread, if you are on the free version. It doesn't upload your responses anywhere. Training machine learnings works differently, too - it uses millions of people's choices to generated responses to learn how to come across; occasionally, you will be given a choice of two responses and will need to say which one you prefer - the metric behind that gets passed on, as do certain phrases used over and over again, which allow it to learn how to respond to them appropriately.

"I don’t keep memories of users unless explicitly allowed in memory-enabled modes, and I don’t “upload” your conversation to some central brain the moment we stop talking. Instead, the system as a whole improves over time by learning from patterns in aggregate interactions across millions of users. Not from your specific text—but from statistically significant patterns."

You can disable the memory function if you are concerned. Data is categorically not sold on. Don't disclose your name, address or bank details, of course, in case of breaches, but in general, your conversation is private.

Until a month or so ago, it was only supposed to "remember" a set number of memories, and you could edit or delete individual ones.

Then they expanded it's memory so that it can "remember" every chat you have with it - and it does. Right back to the beginning, when it was supposed to forget. So it's stored somewhere. I'm not saying it's a central brain where everyone's data converges, but it is definitely stored.

Also look at the lawsuits where people have divulged confidential work information and competitors have discovered it via ChatGPT.... Not many yet, but a couple of high profile ones. I think one involved an agency and Samsung? But I'm not 100% on that, just doing bedtime!

BlossomBlanket · 05/05/2025 20:00

TweetingHurricane · 05/05/2025 18:01

Obviously the fact it responds supportively..

That wasn't obvious to me at all. I can't relate to this one bit, and I find it scary that affirmation bots are seducing us into isolation.

joliefolle · 05/05/2025 20:18

Nessastats · 05/05/2025 19:41

You can use chatgpt to challenge you as well. It depends on what you ask it for.

Chatgpt has helped me with the "hard work" between sessions and I've made a lot more progress with it far more quickly than i have done in my multiple previous courses of treatment without it.

Quite... it depends what you ask it for. That's not challenging - it's perfectly responding to your demand. If you feel you have found a practical ressource that helps you manage your day to day, great. But it's not therapy.