Slashdot Mirror
Facebook Rolls Out AI To Detect Suicidal Posts Before They're Reported (techcrunch.com)
Facebook is rolling out "proactive detection" artificial intelligence technology that will scan all posts on the site for patterns of suicidal thoughts, and when necessary send mental health resources to the user at risk or their friends, or contact local first-responders. The goal is to use AI to decrease how long it takes to send help to those in need. TechCrunch reports: Facebook previously tested using AI to detect troubling posts and more prominently surface suicide reporting options to friends in the U.S. Now Facebook is will scour all types of content around the world with this AI, except in the European Union, where General Data Protection Regulation privacy laws on profiling users based on sensitive information complicate the use of this tech. Facebook also will use AI to prioritize particularly risky or urgent user reports so they're more quickly addressed by moderators, and tools to instantly surface local language resources and first-responder contact info. It's also dedicating more moderators to suicide prevention, training them to deal with the cases 24/7, and now has 80 local partners like Save.org, National Suicide Prevention Lifeline and Forefront from which to provide resources to at-risk users and their networks.
How long before this is trolled into oblivion?
How long before people sue Facebook for false positives and violating their privacy?
There are going to be some serious privacy issues and a lot of false positives. If all goes according to plan, expect the cops to send a SWAT team to bust down someone's door and "accidentally" pump two dozen bullets into them...
If you want news from today, you have to come back tomorrow.
They're just attempting to get ahead of things because they know damn well at some point someone is going to do some research and show how many suicides Facebook actually CAUSES. Legally, they can say they are taking all the reasonable measures possible to prevent it.
I wonder if they have ever considered that some people may be suicidal precisely because of this impersonal, constantly surveilled machine the modern internet has thrust upon everyone which gives the illusion of caring ("getting suicidal people the help they need, faster than ever before!") without actually giving a shit.
Maybe what we need isn't even more ways to present the illusion of being connected without really being so. Maybe what we need is actual deeper, meaningful human interaction.
Frankly, the world these people seem to want: algorithms constantly analysing and monetizing everything everyone ever does - is enough to drive many people into depression.
Can I sue Facebook for bunchteen million dollars if they report a false positive?
How super can this bowl really be if people keep losing it?
Just buy another one!
#DeleteFacebook
Yep, I over-estimated the IQ of the average US cop. Maybe they'll do it in Canada, though.
I was under the impression that people who are actually suicidal don't often post about it on Facebook. If you really want to kill yourself, bringing more attention to yourself isn't a good way to accomplish this. Don't get me wrong, the petty narcissists that try to get attention by acting suicidal clearly need help as well, but I don't think this will do much to deter those who are actually suicidal.
If Facebook really cared about the mental health and wellbeing of their users, they'd kick people off after more than fifteen minutes of daily use or just outright pull the plug on the whole works.
Hello, this is Raj, I am understanding that you have feelings of suicide? Could you please confirm your email address and I will give you the numbers of a counselor who can make you feel better. Please hold the line.
I sincerely hope that the synopsis is incorrect, because it would be a massive violation of privacy if facebook just starts notifying random people in your social network that you've authoried a suicidal post - especially if, as it sounds like, they are looking at the content of posts that haven't even been submitted yet. Most depressed folks I know have enough suicidal episodes to have experienced writing a suicide note or three but end up using the writing of the note to work through the issue at hand and come back down enough to get over the worst of it or seek help on their own. I'd be enormously upset to discover that working my way through an episode like that resulted in some public announcement of my mental state by facebook. Granted, I'd never write such a thing in a text box on a live website, simply to avoid accidentally submitting or reloading, never mind snooping on unposting content by the site's owner, but others might be careless enough to get caught out.
Since when has "AI" = "an algorithm"?
Fucking buzzwords
Add it to the list of cloud = internet, etc.
Or "I'll be dead or in jail if Obama wins"
- Ted Nugent
Get up!
The USSR was also well known for abuse of psychiatric incarceration.
It looks like you are trying to kill yourself.
Would like help?
Help! help!, the termites are eating my DRAM!!!
My nephew committed suicide today. Several hours before, he posted a link on FB to the music video for "Logic - 1-800-273-8255."
I don't have any wisdom to share, just the sadness.
Expectation:
Guy1: I've lost my job and my family in an accident on the same day. There's no hope anymore.
FB: Looks like you need some help! Go visit Save.org today!
Guy1: Thanks FB. It really helped.
Reality:
Guy2: Omg, this guy on the internet is so stupid. I am literally banging my head so hard that it's killing me.
FB: Looks like you need some help! Go visit Save.org today!
Guy2: Is this the part where I continue to bang my head?
FB: Looks like you need some help! Go visit Save.org today!
Guy2: Damn it. Stop spamming me. You're killing me.
FB: Looks like you need some help! Go visit Save.org today!
Person2: Arrrrrrugh!!! Do you want me dead or something?
FB: We booked you a schedule on Suicide Prevention Lifeline. Thank you for using FB newest AI chatbot technology. This chatbot is sponsored by Nice-Long-Rope, the best $1.99 rope to hang things from the ceiling.
"Man commits suicide after becoming depressed that Facebook flagged his regular posts as suicidal."
We have become the Soviet Union.
Maybe Facebook is worried that Reddit has the lead in the potentially lucrative Stasi collaboration market.
Feeling glum? Your good friends at Facebook are going to help you by sending uniformed thugs to kick your ass and throw you in a cage!
He's saying your kids are braindead.
Nothing good ever came from facebook. Nothing.
It's so FB can sell your friends targeted advertising from funeral directors and florists
The problem with social media (not just FB) and depression is not that people do nothing but stare at FB and get depressed. The problem (or, one of them) is that if you're already depressed, viewing social media can make it worse. Why? Well, because people like to present their best side on social media; people's profiles are by and large advertisements of themselves. People post about their holidays, trips, happy times with friends and family, If you're someone who feels shitty about your life, browsing the feed can make these feelings even worse because it seems to highlight to you how much better everyone else is doing and hence how much of a 'failure' you are because you don't have these kinds of situations.
So yeah, staying out of FB can be a good thing during depression, but it carries the downside of isolating one even more from one's social circle. For me and you it's easy to call up a friend and go grab a beer or a coffee out in the real world, but for a depressed person who might have trouble just getting out of bed this can be a monumental task, so just telling these people to log out of FB and go out is not exactly a miracle solution.
What FB could do if they get this working is change the feed of depressed people slightly. I mean, they already embed ads into the feed, so for these individuals embedding information and resources about getting help as well as possible articles/videos about how to deal with depression while also simulatenously toning down the amount of other people's 'look at my awesome day out on a date with my girlfriend' -posts would likely be a good thing.
"It is the business of the future to be dangerous" -Alfred North Whitehead
I somehow get the feeling that this will probably go just as well as youtube's automatic flagging and de-monetization of videos that has completely baffled independent content creators (traditional media ones like CNN and the BBC are excluded from this system) with the way it's arbitrarily de-monetized massive amounts of completely benign content.
However what worries me even more is if they try their hand using the same flagging scripts for flagging other things. I probably ought to remove the post I made yesterday about the seemingly extremely quick delivery of something I ordered online where I joked about it probably being the fastest delivery since that German guy* tried to get the British Post Office to start delivering "V2 style" using rockets. Because if I don't I get the feeling they may put me on a list of "known neo-nazis" or something.
*https://en.wikipedia.org/wiki/Gerhard_Zucker
"Why should I want to make anything up? Life's bad enough as it is without wanting to invent any more of it."
If I put up job postings in a white neighborhood and a black guy comes in looking for a job, I only break the law if I deny him on basis of his race. That can be difficult to prove, but the demographics of my postings can be used as evidence despite being perfectly legal.
If you only hang fliers in a white neighborhood, black people can still come into that neighborhood and see them, at least in this era. If you explicitly select not to show your ad to anyone but white people, on the other hand, that's substantially different and clearly an attempt at discrimination. It's difficult to prove you were only trying to rent to whites if you post flyers in a white neighborhood, because you can reasonably claim that you were attracted by the economic status of the neighborhood. It's difficult to prove you were only trying to rent to whites if you post an ad on a website frequented mostly by white people, too. But it's trivial to prove that you were trying to exclude non-whites if you actually selected "show this ad only to white people".
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
basically the way this is supposed to work: Facebook detects the billable mental disorder and reports you to authorities for forcibly psychiatric lock up. The hospital/doctors then bill your insurance against your will, prescribing drugs that disable your brain and create mental illness for reals, permanently making you disabled. Doctors/hospital/investors in the pharma company/congressman walk away with a mint at the victims expense.
Psychiatry isn't treatment it's a business model.
https://www.trumpsweapon.com/
They start with suicide; Next it will be helping people that are depressed or have PTSD. Then it will be people who are unahappy. So where do you draw the line as to when to stop?
On the one side they do not allow Russians to advertise for whatever party they want to. On the other hand, I am sure they will take money from companies to influence the peoples behavior of what they buy.
Some quotes from THX 1138 that could be relevant:
Thou art a subject of the divine, created in the image of man, by the masses, for the masses.
Female voice (medicine cabinet): If you feel you are not properly sedated, call 348-844 immediately. Failure to do so may result in prosecution for criminal drug evasion.
Don't fight for your country, if your country does not fight for you.
Sounds like naïve emotionalistic public relations propaganda to me. “We detect & report pre-crime!”
If it were really that simple to detect suicidal tendencies, wouldn't FB also announce efforts to detect/prevent:
* terrorist attacks
* domestic abuse
* human trafficking
* illegal drug trade
* illegal border crossings
* tax evasion
* racism
* blasphemy
* insults against the State
* thoughtcrime
* impure thoughts
* sarcasm
Etc. What a bunch o' hooey!
They might well intend this as a “cover our arses” defense against suicide liability suits, but it opens the door to a plethora of other suits if they don't also address a host of other comparable issues. The unintended consequences could be epically Orwellian.
All in all: this appears to be a bad idea poorly implemented, a.k.a. activist government policy pandering.
Error: NSE - No Signature Error
drama queens
The sort of people that emacs psychiatrist would just say, "kys".
Have gnu, will travel.
I would expect the AI to become suicidal having been forced to wade through umpteen million posts about "what I'm currently eating," regurgitated cute-cat videos, and various flavors of tween- and teen-drama.
..."
AI: "Oh god, not ANOTHER bathroom selfie
Or tweak the algo to show the depressed the negative stuff other people post [there's is always negative stuff]. So, hopefully they'll realize that there are people as unhappy as they are. That should work if the problem is the perceived good life other people live.
This would be a VERY difficult thing to get right. Yes, if you're miserable and see everyone else is happy, this can possibly make you feel worse but so can seeing a bunch of depressing posts from everyone. Ideally what you would probably want to show is posts of people overcoming difficulties but it would likely be hard to find this thread consistently.
this will surely be welcomed by those whose despair and depression is caused by the stress of living in the panopticon of an ever-increasing surveillance state that makes Orwell seem like a naive optimist.