[Dark] Depression/Suicides within the fandom (The revival)

Facebook now is going to monitor our postings if we present suicide trends. Its use is prohibited in the European Union

@Background Pony #A051  
I’m guessing what they mean by “Help them get support quickly” is “call the police covertly so they can forcefully intervene”. Facebook is going full 1984, as such I have already migrated to minds.com

@Genevamode  
I mean, thats what there doing beacuse false or misleading calls to 911 that the algorithm got wrong would cost millions.
 
But, I do like the concept. And no, it’s not 1984 beacuse, unless privacy settings are set up, anything you post on Facebook is public.

@warc9  
You’ve got a point, but it still doesn’t sit right with me.

I was at work and hit everything and thought I switched it right back and then accidently hit random and a questionable image popped up lol
 
Could have been worse, I guess… But now I feel like I’m looking at porn at work

@Background Pony #A051  
I would rather just have a message sent out to campaign for preventing suicides instead of having it monitored by AI. I wish that they would just make a suicide comment-flagging tool or something.
 
I imagine that the software would cause a lot of false alarms as people say stuff like “This makes me want to kill myself” and similarly-worded phrases.

I wish I didn’t have autism.

@Background Pony #A9C7  
Amen. It’s a curse.

@Genevamode  
I wish I could cure myself and feel like I belong to this world.

@Tranquil Night  
Actually, the Ai aren’t going to call the authority figure for help. Human m0derators also involved too, to prevent the false positive alarm.
 
“It’s the human m0derators that will do the crucial work of assessing each case the AI flags and responding.”
 

 
 
https://www.theverge.com/2017/11/28/16709224/facebook-suicidal-thoughts-ai-help

@CyanLightning  
Oh.

I always feel a mix of emotions

Ever find yourself repeating a random word hoping it’ll help

I feel happy but I know I could be a lot happier if I did certain shit that I’m too lazy to do… like move out of my fucking parent’s house… or get more hours at work… and just do any other task that is… I don’t fucking know… it’s like.. once I do it I’m 100% glad I did it… but to actually do it I have to push myself just a liiiittle.. it’s not even hard… what the fuck is wrong with me… like a lot of these tasks I’m talking about AREN’T FUCKING HARD AT ALL… yet… I just don’t fucking know…
 
And I keep having thoughts that something bad will happen… I don’t know why… probably because the world is fucked up and shit can fuck you over really quick

I often feel like people see me as a clown that they feel free to beat up on.

I had depression. But I handled it. Now im depressed a bit. Just because im bored. Its a dark room. I like dark. I shouldnt focus to bad feelings.
 
I like electronic music producing, programing, drawing, but now i dont have the mood to do anything.

I’m sitting here, I’m on Discord, yet I still feel lonely as fuck… Maybe because I like the booru more… Friends are inactive… fuck… fuck… FUCK!

@Dave33333  
Go to the mall and kill some time by window shopping.

@warc9  
If I was old enough to have a drivers license/actually have a mall near by, sure.

@Dave33333  
Oh.
 
Strip mall?

@warc9  
Closest thing to a mall here is literally a Wal-Mart.

@Dave33333  
Well, I spent alot of time in a walmart looking at stuff today.
 
Go look at the x mas decorations, some of them are cute.

@Tranquil Night @Genevamode @warc9  
Anon here, sorry to answer late.  
I’m going to try to explain this latest new if you didn’t know these details: The artificial intelligence analyzed the health history of more than three thousand people who attempted suicide, understood in detail the combination of factors involved and predicted future suicide risks; He managed to accurately predict 85% of future suicide cases two years in advance and had a higher accuracy with 92% when it was a week before the attempt.
 
I have no problem for an artificial * intelligence * to examine my life on Facebook, and it would not bother me much in person, especially when i think it is safe and I can benefit, but yes, the percentages aren’t 100%, with 15% or less in fail, but with what @CyanLightning said, i don’t think i should worry too much.  

 
@Dave33333  
I know that feel “sniff”. im so lonely ;-;

@Background Pony #A051  
As long as humans are involved in the final determination of whether or not the police are involved, I am fine with the system.
 
Artificial intelligence makes mistakes in analyzing the speech patterns of humans; it can be mostly accurate, but it will not be as accurate as humans are. Humans make mistakes as well due to their emotions. Even if a human correctly identified a soon-to-be suicide victim with enough time to flag the comment and have the police notified, many people would refuse to do this due to their emotions.
 
I prefer humans working with machines; the AI (and I believe that humans should be involved in this as well) can flag the comments for humans to look at the information, and humans will ultimately call the authorities.
 
However, I would not be surprised if that fact was twisted to make it look better than it was. If the AI analyzed 1000 nonsuicidal comments and flagged “suicidal” for 80 of them a week before a given date, it would still have an accuracy of 92%.

@Tranquil Night
 
Artificial and human intelligence differ greatly, I don’t doubt it, but they are tools and their usefulness will depend on us, that is why it is normal for them to fail but it is not a reason to believe that they can not be as precise as humans at some point.
 
In this case yes, the IA needs human supervision, having 85%-95% is not enough evidence to leave her alone in a job as ambiguous as this one of how a person feels in wanting to take his own life.