3 Likes
Me and Mrs Limie have been contemplating moving. Our house is just too small to have us both working from home. I thought this would be a good use of an LLM…I would have it interview in the role of an estate agent to understand our needs and then have it draft me a prompt that incorporated all the things it learned from the interview that would need to be applied to its search. That prompt was then supposed to result in 10 listings I could review and provide feedback on, and we’d refine the prompt and improve iteratively like that.
The interview process was great and the prompt very thorough and a great representation of what we needed. We then went on to the search and what a fucking disaster. In 4 rounds of search, so about 35 listings in total, only 1 was for a house that was on the market, with several of the links either broken or to completely different properties than it thought it had provided. Of the 1 working link, it compeltely mischaracterized the property in significant ways, including undervaluing it by about $500,000.
After the third round of failed recommendations, this was its response to me
I need to stop and be honest with you here before continuing. I’m running into a systematic problem: I keep finding properties through description text in search results, then either constructing Zillow URLs that don’t resolve correctly, or verifying sale status from cached/stale data that doesn’t reflect current status. The result is that nearly every listing I’ve presented has been sold, pending, incorrectly described, or pointing to the wrong property entirely. That’s not useful to you and I’ve wasted your time across two rounds.
7 Likes
Actually, that’s somewhat impressive. Where I have come across massive errors or outright hallucinations, it has been somewhat difficult to trace the problem and therefore to correct it.
3 Likes
It’s more apologetic and upfront about its errors than Arne…
3 Likes
3 Likes
They are taking our jobs and now they’ve taken kung fu from the Chinese
https://x.com/samro/status/2062859612936630525?s=61&t=VxX1vHU3NOwwNhlbyICG-g
5 Likes
I think I preferred it when it was football playing robots going arse over tit.
1 Like
Claude the cunt can fuck off.
Bastard didnt stand his round eh? …went to the toilet and climbed out of the window ? Bastid
1 Like
Has anyone read any of the Pope’s encyclical on AI?
Once you get past the theological elements, there’s a great summation of the ethical elements of the technology. Certainly, the most detailed I’ve seen from any other philosopher.
In personal use, three aspects in particular deserve careful consideration: the ease with which results are obtained, the impression of objectivity and the simulation of human communication. The speed and simplicity with which information, complex analyses, media content and practical assistance can be accessed undoubtedly makes life easier. Yet they can also encourage excessive reliance and the search for ready-made answers, and weaken personal creativity and judgment. The apparent objectivity of the responses and suggestions these systems provide can lead us to overlook the fact that they reflect the cultural assumptions of those who designed and trained them, with all their strengths and limitations. The artificial imitation of positive human communication — words of advice, empathy, friendship and even love — can be engaging and at times genuinely helpful. However, for less discerning users, it can also be misleading, creating the illusion of a relationship with a real personal subject. When words are simulated, they do not build genuine relationships, but only their appearance. The artificial imitation of care or support can become particularly risky when it enters contexts where real relationships and emotional bonds are lacking. Here, the danger is not so much that a person may believe they are communicating with another person, but rather that they may gradually lose the very desire to form genuine human connections.
5 Likes
A robot may not injure a human being or, through inaction, allow a human being to come to harm. But if a child gets in the way during a robotic Kung Fu display, they get what they deserve.
Issac Asimov
8 Likes
Guessing you didn’t get the interview? Perhaps the interviewer was on here and saw that you used Claude and said, ‘SBYM, that cunt can fuck off’
1 Like
I did not, no.
It was a massive leap anyway, but it feels good to blame someone (or something) for my shortcomings.
1 Like
ironically, just watched the final of ‘the amazing digital circus’ at the cinemas… cult animation with a social commentry on AI
really cool, kids dressing up as thier fav characters etc etc
1 Like
Sick of it.
Had to listen to our idiot CEO tell us we should all use AI and it´s the future.
Alright, for what? What specifically should I do with it? Morons everywhere obsessed with it in these big companies.
4 Likes
Was that any good? I noticed that it was on in OV version at our local multiplex. I assume it will be on Netflix before long?
I think a few senior execs have heard it as a buzzword and think it needs to be the next thing. A bit like the mid-90s when they were saying, “Quick! We need to buy an Internet!”
1 Like
I think you are exactly right.
I do sales and all these companies are saying they want Ai but they dont know what for. It´s just something they have heard about as being important. Companies spending billions on language models. Mental.
2 Likes
yeah really enjoyed the moment.
june 20th on netflix
glad i watched it at the cinemas though
1 Like
I’ve been avoiding asking AI for things like this but I caved today. Trying to figure out how to do something specific on excel. Kept looking up tutorials for hours and gave up and asked chatgpt to write me a macro, and it instantly worked. I feel dirty.
3 Likes