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?

Or judging someone for using calculators.

I'd judge someone for using calculators if calculators were demonstrably incorrect well over half the time

I love it. It gives me better advice than any human

It has definitely made life so so much easier as I write copy for work and often need a second opinion.

If you can't see that you are setting yourself up to be replaced entirely, then I can't help you.

Well @snowmichael quite a lot of assumptions in your reply there. One could say that to make those assumptions is lazy and stupid. Your reply is also patronising and rude.

I said I used a lot in work. At no point did I say that I just cut and paste or present work written by bots.

AI is a co-pilot not an autopilot. It is useful for idea generation, analysis and many other functions. It still needs a human brain to understand, contextualise and make decisions about what is useful and what is not.

Well, that's what computers did. Think about the pace of work now compared to when a letter was dictated, typed out, waiting for a reply, etc.

ill admit its much easier for most of things, although does anyone know which ones are good for research eg historical etc ?

None. They lie. In academic matters particularly.

They will confidently give you bullshit facts backed up by entirely made up references, as in they will cite works / papers / books which don’t exist.

Funnily enough I was reading a book this week about a book being published in the 1930s. I thought it would have taken so much longer than it does these days but guess what? Turns out the postal service was so much better with multiple deliveries a day, it was just as fast as it is now.

heres an example of one of the Finland data center companies using the excess energy from the growth in data centers, much due to AI growth for sure, for environmental benefit. There are lots.

https://www.equinix.com/newsroom/press-releases/united-kingdom/2024/10/more-helsinki-homes-to-be-heated-using-excess-heat-from-equinix-data-center

Be part of the solution people, not just a virtue signaller claiming you are not part of the problem because you only fly once a year and only use the internet for important things.

You can ask your AI tool to not withhold data or train their models, and use a temporary chat so that it deletes it after you're done. The best way to be GDPR compliant however is to use a paid subscription, which I suspect many don't.

I literally agree. It’s not a good thing. I just meant that it will be unavoidable - for example how we’ve been forced into a position where you essentially need a phone to do basic things like banking because the business just won’t deal with you otherwise and you have no other options. In the case of AI, you will still be engaging as every business will use it. So if you catch a bus which uses AI to process tickets, you’ve contributed to the use of AI (against your will, perhaps, but you have).

The post, yes, but the work inside the office was slower. If you made a mistake in a letter that was too big to be fixed with tippex, you had to re-type the whole letter.
I suppose also that they would have used a phone for many of the things we now use email for...

Same. My students are not using it critically but consume the bland, utterly meaningless corporate good feelz solution to every problem (deliberately engineered to produce) without a second hesitation. You need baseline knowledge to critique its output AND a critical mindset. Some colleagues are trying but many are like those on these threads, focused on the convenience (and it is hugely convenient) rather than implications.

And you know these guardrails are upheld how?

Are you relying on a firm that knowingly broke every copyright going and is working through thousands of lawsuits…

In the 1930s you'd just XXXX out the typo and keep going. But I do agree with you that expectations of output are rising, deadlines getting shorter, more churn etc. As I said upthread, one of many problems is that it's supposed to make the workplace more efficient but it just creates more work further down the line. Companies are getting rid of people at step A in the workflow to replace them with AI, then creating more work checking at step B but not hiring extra people to do it.

e.g. Duolingo has fired all its in-house linguists to move over to AI content creation. It is now hiring a handful of externalised linguists on short-term contracts to clean up the AI mess, with much worse pay, conditions and job satisfaction.

And that's aside from the whole situation where we urgently need to degrow the economy so our children don't all starve to death in the great crop failures of 2043.

I don't know. But I trust it the same way I trust phones aren't actually listening to us even after I turn off Siri, or I trust that I'm taken off a mailing list when I request it. Not sure what else one (who is choosing to use AI) can do.

I read a thread on Reddit where someone unwittingly bought an AI-generated book on Amazon. It was a guide to mushroom foraging and he poisoned his entire family (thankfully not fatally). Imagine the liability chain 😱

I don’t really fine AI useful at all.

Tried to use at work to write code and it does it very inefficiently. Tried it to write some bits for work but what it came up with was really formal and not the right tone.

I’ve tried it for holiday stuff we have an 18 month old and everything is suggested wasn’t suitable.

The most useful thing it’s done for me is tell me and link me to episodes of podcasts I want to listen to without having to scroll through episodes. Although this takes a while sometimes as it links the wrong thing…

This is both co pilot and chat gpt.

For the things OP mentions and other information overload reducing tasks, AI is common sense. I might try it.

For the writing and emailing, customer service, school and general communication with other humans I think it's harmful and dangerous and will never use it.

Sounds like that's AI doing what it was made for, i.e. to help out.

No problem with that.

It's where we get too reliant on something, tech or otherwise, that we lose the ability to do it ourselves that it becomes a problem.

You learned to make sure that you didn't make big mistakes in a letter. Yes, we did use a phone for many of the things we now use email for, and could ask all the questions and get the answers in one call, rather than emailing back and forth to clarify things.

As an admin worker I found office work far more fun and rewarding in the days when you actually had to use your brain and do things manually. Computers killed any joy I got from working in an office, and if I could go back in time and knew what I now know I would have done something more interesting.

Copilot has been absolutely brilliant for me as it's enabled me to make pretty complex spreadsheets and databases, far beyond what I'd previously been capable of from conventional courses on Excel and Access. I can give it exact details of entries in specific cells/fields and tell it what I want in other cells/fields and it pops up with formulae etc to use, which most of time do exactly what I want. If it's suggestion doesn't work, I just tell it that it doesn't work, give the example why it doesn't work and it comes up with a better formulae!

I'm very late to it. It was my son who told me about it. He works for one of the country's biggest insurers and they used it daily when designing spreadsheets and databases for annuities. He said that previously they'd have to make formal requests to their IT departments for "one off" spreadsheets or databases that could take weeks or months to be created, but now they can do the same within a few hours.

He showed me how it works and I've been a fan ever since.

I’d suggest that a mailing list is not wholly comparable. But agree that there’s not much that you can do other than assume your inputs and prompts are not confidential and that you are giving away more data than you realise.

I don’t.

DH’s job involves AI and although he said it’s very helpful he’s also very worried about it. Building machines that gather data and can think for themselves is a real threat to the world as we know it.

Even new Chinese cars will be gathering data on location. They’ll be capable of shutting down cars en masse shortly if they can’t do it already.

It’s like the internet. Great in lots of ways but increasingly becoming dominated by the dark side of what you can access in terms of porn, abuse, violence, hackers, atrocities, etc.

i've used it for some therapy, i have had LOADS of CBT for anxiety, but this morning it helped me stop spiralling about something and centre myself about what i was struggling with.

Do i think it's the best therapist in the world? Absolutely not, did it help? Yes.