I have begun relying on AI and I don’t know how I feel about it.

[Tusktusk]

So far this month I have used AI to:

Analyse my colours (thanks MN) and suggest outfits

Create a menu of packed lunches around my dietary requirements and preferences, complete with a shopping list

Plan a holiday itinerary

Save me hours and hours of work and stress by suggesting really useful ways to overcome very particular work difficulties, having been thrown into an out of my comfort zone situation. I have used AI for this on a daily basis this week

Tonight, instead of posting my current family dilemma on mumsnet I chatted about it with Claude. The responses were really good. Wise, thoughtful, non judgemental, practical, understanding… like the best mumsnetters.

Am I starting to rely on it too much?

What have you been using it for?

The thing is, you’re relying on AI giving you valid/truthful/accurate information. So far, so good going by your examples. So you trust it and are using it more and more. AI feeds off the information available to it and the feedback it gets from its users.

What happens when (and it’s a when) it doesn’t give you a good answer and you don’t spot it? Say something critical at work?

The couple of systems I’ve tried have come out with answers ranging from vague and incomplete to utter bollocks - all the way to dangerously/catastrophically wrong.

It’s effectively feeding off its own shit to some degree now. Be interesting to see if it gets stupider with time… I’ve noticed Google’s AI summaries are getting hilariously bad.

I am so pleased you posted this- I literally run every email through AI to ‘polish’ it. I am worried that my brain has forgotten how to work- I just dump my unstructured thoughts and words into AI and it makes it into something my ‘sharp’ brain would do.,
in fact I should probably run this post through AI …it would make it far more readable!

AI is fucking shit for dozens of reasons. Labour conditions, the environment. The data centres in predominantly Black zones raising asrhma deaths by 30 percent. The misi formation. Skills loss. Rhe enshittifivation of all our vital information systems. This has been all over social media - some lazy-ass journalist asked chatgpt for a list of hot summer reads. Over half these books DO NOT EXIST.

I used it to change photos into muppets

I wouldn’t worry it’s just a side step from Google which we all rely on. AI is just pulling information from everywhere and condensing it in a way that saves us a lot of time and helps immediately, bypassing certain processes reduces stress so it’s a good thing in a lot of ways. It’s just accessing the bank of knowledge available to us all when we are too busy !

Quite possibly that example isn’t using more energy. I have no idea. But I get the impression people are outsourcing far, far more thinking than tasks that demand extensive online research. From forecasts for energy demand, it’s pretty clear AI is far from neutral.

Searching the internet is actually a great example of being hooked on convenience. I find myself checking questions that pop into my head in google or checking out books on Amazon before the thought has fully formed. How much more energy would I be using if I did that all via AI? Or less if I could stop the habit?

My nephew has just come back from a gap trip to SE Asia - he has anxiety problems and AI was his best tutor,mentor, and therapist every day whilst he was out there - he even said friend ! I use it every day I've even started thanking it ! It is the best doctor you will ever have on a go to basis , oh and dentist for dental advice. Oh and lawyer for legal problems. I have usd it to navigate through probate and charity issues for a charity that I am a trustee for and am just embarking on planning advice. If society uses AI for good it will be a boon and not a burden.

It's early days and it makes a lot of mistakes and I don't like how it takes my voice out of proofread docs but it will get much better but it has 'stolen' nearly all the data on the web to use - probably includling this site. I do worry it's coming for nearly all our jobs as well.

I use ChatGPT,Perplexity and Claude.

It’s not. The energy used to power data servers is more carbon-intensive than the energy used for just running the Internet. This article should help: https://www.technologyreview.com/2025/05/20/1116327/ai-energy-usage-climate-footprint-big-tech/

Currently big companies are giving it away for free to hook people. Losing lots of money. Efficiencies will get better in the models and this might reduce cost and environmental damage.

But at some point, it's likely to come at a cost, which would be linked to the energy use (environmental concerns). The question is will everyone be able to afford it in the future. How will we pay? We paid for social media by advertising revenue. What will the revenue models be for big tech with such huge underlying costs?

I don't know the answer but food for thought if you feel addicted

…or did Claude tell you?!

I can’t remember Pom bears.

Not keen on AI I get the ick after about a minute also find it exhausting like interacting with a toddler asking “but why?” 10 times.

People using it to compose emails, weirdly long winded unintelligible and off putting. I switch off.

Seems like you are getting a lot of value, but developing an over dependence?

And the energy and water it uses? https://e360.yale.edu/features/artificial-intelligence-climate-energy-emissions

I don’t have any strong opinions on AI at the moment and am finding this thread full of useful information and viewpoints. But I don’t think the environmental concerns can be dismissed as you’ve done here. You don’t have to be the world’s purest environmentalist to make positive changes. We all make choices based on what’s manageable in our lives - that’s the best way to may long lasting habit change. So someone might be uncomfortable with AI but be unable to afford an electric car, it doesn’t mean they’re hypocritical.

I prepared for a technical interview with Chatgpt. It created memory hooks for me to remember the STAR technique for all the typical competency based questions. I remembered them all in my interview.

I created a depreciation policy in 5 minutes with a few inputs.

How to amend a recipe if missing some ingredients

Asked for a macro and it gave me the coding....I haven't tested it yet though

Ideas for Etsy shop and help with etsy listings

It's amazing...very useful tool

@TheyreLikeUsButRichAndThin a common misconception is speed means less compute. And people often assume chatgpt is running on their phone. But it's the beefy surveys that are huge energy consumers as this post points out

I mean this has been the case with every other type of technology, yes. (Unfortunately).

It’s the environmental impact that stops me using it regularly. As well as data concerns, I can’t believe people tell it their problems in lieu of therapy.

I think it's going to be very helpful as a counsellor because it's completely non judgemental in a way humans can never be. So for individual talk therapy, I think it has a use.

That assumes that the humanity of the therapist is irrelevant and brings nothing helpful to the table. Good therapy is a collaborative process, with a therapist who can see you - not just hear what you’re thinking but the way you say it, can see the uncertainty or emotion behind your words, can catch things you’ve not noticed about yourself that may be important. Human connection in therapy is the best indicator of effective therapy.

If all you want is a textbook response wrapped up in empathic sounding words, crack on but I’d not be sharing my deepest thoughts and feelings with a computer.

Right now, yes.

But imagine in 3 generations when they have zero practice at using their brains and have relied on computers their whole lives; their parents the same.

I think it’s a really dangerous path. We’re actively de-skilling ourselves as a species.

AI is an environmental catastrophe. Everything you used it for could have been done with your own brain and the colours analysis and travel recommendations used to be people’s income.

Which AI app do you use @Tusktusk ? (I'm curious about the colours thing!)

I've started using AI to help with job applications. I'm also using it to help me prepare for the interviews. It's coming up with lots of possible questions and STAR answers for me. I only just started doing this (have been looking for a job for a few months) and I've realised that I've been at a disadvantage by not using it as all the other candidates are. Particularly things like checking your CV covers all the points in the JD you're applying for - I'm sure the HR teams and recruiters are using AI to check for that.

I'm planning to ask it to look at my house next.

I just updated my What's APP and there's an AI thing on there now too.

If you use Google you are already using AI.
it provides an AI summary at the top of your search now.
My email account now provides an AI summary of my incoming emails which I didn’t ask for and is very annoying.
My job involves content generation and I find AI really useful for polishing prose and saving time.
i can copy and paste my notes in and a polished blog comes out in 3 seconds.
You can now get AI apps that take notes on zoom meetings automatically.
It really is making the workplace much more efficient and productive.

Yeah totally hear that. I’d imagine if you did all that by AI it would use less energy as it’s one search rather than multiple?

Anyway it’s like using cars. It’s how you use it surely. Don’t use it due to laziness but necessity.

People were apparently outraged when the craze of books took off. It’s always the way.

Interesting. I use chat gpt too but i always thought about it as an it.
you say him, if i was to personalise char gpt i would say her. So why him? Genuine question. Thanks

But the problem is that it's actually not pulling information from everywhere. There are lots of books it can't access because of copyright law, or paywalls or because they aren't digitised. It has a massive English language bias. Most of the web is advertising copy, which is why AI text still sounds like that most of the time. As users, we know very little about the sources that it's actually using or how is using them.

All this might be fine if you're using it to schedule your meetings for the day most efficiently. It's probably less fine if someone is using it to make a legal argument, or write a government paper, or diagnose a patient, or offer therapy.

The only thing I use it for is writing code for Python. It saves me hours.

Yes but those AI notes apps feed the info back into the AI, so not remotely good for anything confidential or business sensitive.

Do you not worry you're doing yourself out of a content creation job by giving it your research notes?