36 minutes ago
>Thackray’s lawyer, Danilo Rajkovic, told the court his client had gone on a stress and frustration- induced tirade after trying to use the Claude platform to help treat an ear infection.
AI is not your friend and neither are tech companies. Repeat that as many times as you need to until it is fully understood.
6 hours ago
7 hours ago
I know the argument of "well what if he did it", but this is chilling.
6 hours ago
I'm trying to figure out why anyone would have a concern with this.
Seems there's a line and this is well past any reasonable line.
6 hours ago
Well, here we have a man who frustratedly ranted at an inanimate object and was probably surprised to learn that the inanimate object in question has handlers who felt the need to read those rants, and pass them on to Interpol, probably with the intent of having our ranting man arrested.
My dog has never done that to me, so I'd call this story novel and concerning.
5 hours ago
In 1980s Australia a man or woman seen ranting and raving at an inanimate object like a traffic light or hat rack could very easily (but not always) be reported, arrested, and sent for a mental health assessment.
In that context this is different but not novel.
Society is right to be concerned with how such people are assessed and managed, both for the sake of the individuals skirting the edges and for the wider community that might have to deal with larger consequences.
5 hours ago
Where that hat rack is in my study, I have every right to assume it will keep my confidences.
And where that hat rack is the object of my ire, I can have confidence that I won't find myself on the wrong end of a commonwealth law regarding the use of a telephone.
He could get five (5) years for this, because of the 'using a carriage service' element. Tell me you've never uttered oaths or threatened your lawnmower when it won't start, and ask yourself if Husqvarna ought to be a mandatory reporter.
Seems pretty novel to me.
4 hours ago
> Seems pretty novel to me.
Yeah, well, that's a function of life experience then I guess - I'm approaching 70 and I've seen a few people subjected to compulsory assessment (neutral pyschs interview and determine how stable a subject is) and some of those sectioned for a week of observation before determining treatment.
That, from the report, is very much what is happening here - court recommended determination of how cray-cray the person in question is.
Seems like a prudent course of action, to just check at least, if some one ranting at their lawn mower, making threats toward children, etc. is alright there and whether they might want to touch grass a bit.
Where the contention lies here would be in the question of to what degree ChatGPT / Claude et al. are your own confidentiality bound therapist slash priest combo.
Spoiler alert (of course), they are not: https://www.youtube.com/watch?v=Tn4P4m-B8LY
( Former AU police officer turned prosecutor turned public defender opinions on what's actually happening when your AI chats get pulled into a legal case )
5 hours ago
In the article it says he was annoyed with Claude because it wasn't solving some other problem that he was asking it about.
Reading between the lines, it seems he may have got annoyed and made a wrongthink joke to a computer and eventually got arrested for it.
A joke with a computer, no matter how distasteful, is not past my reasonable line.
Getting arrested for making a joke in your own home is, however.
Of course, it stands to reason... if he is charged with planning to carry out a shooting, logic dictates they found a gun? Or plans to obtain one? Bullets?
Or anything at all other than an email from an offended slop generating plagiarism machine?
5 hours ago
i think that argument is a red herring here, wasnt the guy just fucking with the LLM? to me this reads like a bullying version of "my sick grandma" LLM manip
even reading the title you go in with a trama/rage before knowing. brokenmachine's take in this thread is probably spot on
13 minutes ago
Wasn't the kid in Tumbler ridge just fucking with a LLM before he decided to kill other students?
I don't care if the guy threatened a lawnmower, I do care if he started spouting off about killing kids.
Context matters, but that's what the evaluation is for, not you our I deciding that a rant with potential violence towards children is just a rant at an LLM, or a premeditated act in advance of a violent crime.
I don't know. Do you know for sure? What happens if you're wrong?