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 internet will soon no longer be a source of reliable information.

Rip The Internet 2025

well you would use a medical textbook to see if Ai matches etc

i was not debating the finer details my comment was the general advancement of technology, eg why invent the kindle when the book is just as good etc

what is the fucking point of that? just doubling up on labour to check if the bullshit machine matches approved authoritative sources

I mean ... I can take 100 books on holiday for less than the weight of a single paperback, read in the dark, have instant access to any book I want ...

.

and when researching any topic, originally it was go to the library, seach all the books for suitable ones, then read the index's to find the topics, then if needed go to other librarys etc. to do more research on a topic,

now with Ai, the topic in seconds, the information available, then if you already know about the subject but needed extra info then you have it etc

would you trust your friend to tell you about x topics or do you 100 trust them and dont research any of it or get other advice ?

I would trust my friend who had proven expertise in foraging. I wouldn't trust some randomer I don't know. I especially wouldn't trust some grifting bullshit pulled out of the arse of an AI. HTH.

but then you could debate where the friend got their info from etc

The topic in seconds but it is incorrect up to about 60 percent of the time. Can you seriously not see the problem with that?

yes that its an assumption that its incorrect 60% of the time, if people want accuracy then it needs to be peer reviewed studies on how accaute the latest models of Ai / chatgpt etc are

that said there is these days wikipedia to also use as a guide when researching so between the both of them it gives more accuracy.

however if the research needs to be grade A qualiy then it does need to be academic textbooks etc

We have LITERAL CENTURIES of a social contract vesting trust and reliability in print and libraries. Now you want to replace that with a private equity instaresponse black box that demonstrably makes shit up.

yes because in many respects its like tapping into the minds of some of the greatest people after all the Ai's were trained with those books. yes it also makes up information but then so do some programs on tv eg part fact part fiction yet people still believe some of the fictional parts as truth etc.

overall i get what your saying but due to where the books are ect or if cheap enough to get a one etc then in place of that sometimes we need the software for the research

I used AI to help me deal with somebody I discovered is a convicted sex offender. Maybe as life becomes increasingly complex, AI will be the next step in our evolution, sort of. We're not exactly doing very well on our own...

Just to clarify, OpenAI did not train its model on the classics or on published research, it did train on some corpora texts as well as the entirety of Wikipedia as well as Reddit posts. Latterly we came to understand that they’d broken copyright to scrape data off many websites but the model is not providing you with ‘information’ or ‘knowledge’, its responses are entirely based on probabilities. Like a very fancy predictive text on our phones guessing the most likely next word. Word by word. It has developed its algorithm incredibly quickly once it was able to continuously train on 100+ million users.

But it most definitely is not tapping into the minds of the greatest people / thinkers. Personally , I find it incredibly generic.

https://www.ibtimes.co.uk/chatgpt-falsely-accused-me-sexually-harassing-student-law-professor-says-accusations-were-1724935

Did you find out from AI? It has incorrectly named people in the past.

No, I found out they were a convicted sex offender, then used AI to learn about the offenders register, sexual harm prevention orders, likely restrictions and conditions etc before contacting the police and letting them deal with it.

How ChatGPT is trained - generating responses from data - PC Guide

https://www.abc.net.au/news/science/2023-11-29/artificial-intelligence-ai-training-datasets-copyright-books3/103157980

overall why are some people so opposed to using Ai. but happy with other technological advancements ?

overall why are some people so opposed to using Ai. but happy with other technological advancements ?

Because other technological advancements - such as a Kindle, for example, as mentioned above - don't spout bollocks.

I don't trust Wikipedia either, though it's better than it was. I do use the links and references as a starting point, along with Google, if I want to read up on something I don't know a lot about.

In my own subject, I've got a pretty good idea what's a reliable source and what is crap, and I can apply that to other subjects. But the vast majority of people won't know the difference between a piece of AI and a peer reviewed article in an academic journal.

And as for using AI to do your writing for you - well, who needs a brain of their own anyway?

Anyone remember the song 2525?

I asked ChatGPT todays wordle answer after doing it and it couldn’t even get that right

AI is more problematic than other technologies because of its excessive energy and water consumption when we should be dialling way back on those to achieve net zero. It is built on plagiarism, as the link you posted shows. It is proven to erode critical thinking, and it is very easy to use it to spread disinformation. That is why it is bad.

Like all new things - use with caution. It can be useful but check the results to be sure they are true. Don't put anything confidential into it unless it is a work paid for AI with confidentiality built in for what you input.