#crisistextline — Public Fediverse posts

@ai6yr @cindyweinstein I’ve been a volunteer crisis counselor for the #CrisisTextLine for two years, though I’ve been inactive since the spring due to my life getting somewhat chaotic. I’m getting back into it as soon as I can. They need all the volunteers they can get.

@ai6yr @cindyweinstein I’ve been a volunteer crisis counselor for the #CrisisTextLine for two years, though I’ve been inactive since the spring due to my life getting somewhat chaotic. I’m getting back into it as soon as I can. They need all the volunteers they can get.

I'm realizing that #AIDefeat & #NoConsent could be effective tools w/ #Solidarity to resist tech companies bringing "AI" & surveillance to care systems for emotional distress, such as they are. These tools combined with organizing and support to empower people seeking help, who are vulnerable.

It's a paradox, wanting to avoid putting any burden on persons at their most difficult moments. This is also the moment of exploitation for their data and when meaningful consent is not possible for any complicated ask.

This is creative work, these tools could have power of their own. Hopeful.
#MentalHealth #CrisisLines #988Lifeline #CrisisTextLine #StopOxevision

I'm realizing that #AIDefeat & #NoConsent could be effective tools w/ #Solidarity to resist tech companies bringing "AI" & surveillance to care systems for emotional distress, such as they are. These tools combined with organizing and support to empower people seeking help, who are vulnerable.

It's a paradox, wanting to avoid putting any burden on persons at their most difficult moments. This is also the moment of exploitation for their data and when meaningful consent is not possible for any complicated ask.

This is creative work, these tools could have power of their own. Hopeful.
#MentalHealth #CrisisLines #988Lifeline #CrisisTextLine #StopOxevision

2/ [#CrisisTextLine job announcement, link at very end, more where this came from]

Staff Data Engineer.
"Crisis Text Line's software engineers develop the next-generation data platform that changes how millions of humans connect with volunteer crisis counselors, our clinical staff and the interaction with one another. Our platform needs to handle information at a massive scale and extend well beyond SMS as a channel. We're looking for engineers who bring fresh ideas from all areas, including information processing, distributed computing, large-scale system design, networking and data storage, security, artificial intelligence, natural language processing, UI design, and mobile. .."
(cont'd)
#LLM #AI #CrisisLines #DataEthics

2/ [#CrisisTextLine job announcement, link at very end, more where this came from]

Staff Data Engineer.
"Crisis Text Line's software engineers develop the next-generation data platform that changes how millions of humans connect with volunteer crisis counselors, our clinical staff and the interaction with one another. Our platform needs to handle information at a massive scale and extend well beyond SMS as a channel. We're looking for engineers who bring fresh ideas from all areas, including information processing, distributed computing, large-scale system design, networking and data storage, security, artificial intelligence, natural language processing, UI design, and mobile. .."
(cont'd)
#LLM #AI #CrisisLines #DataEthics

1/ What do you think of when you hear "next generation data platform" "information at a massive scale" "extend well beyond SMS as a channel" "artificial intelligence" "natural language processing" "data consumers" "data pipeline development"?

Did you think Crisis Text Line? #CrisisTextLine

It would seem risky for #CrisisTextLine to go all-in on conversation data & “#AI ” tools after being called out 2 years ago--considering the problems with consent. So many children. So many people, vulnerable moments. Still unregulated, except maybe #FTC consumer protections? It's not easy to learn what CTL is doing, so I went to job announcements.
(cont'd)

1/ What do you think of when you hear "next generation data platform" "information at a massive scale" "extend well beyond SMS as a channel" "artificial intelligence" "natural language processing" "data consumers" "data pipeline development"?

Did you think Crisis Text Line? #CrisisTextLine

It would seem risky for #CrisisTextLine to go all-in on conversation data & “#AI ” tools after being called out 2 years ago--considering the problems with consent. So many children. So many people, vulnerable moments. Still unregulated, except maybe #FTC consumer protections? It's not easy to learn what CTL is doing, so I went to job announcements.
(cont'd)

I've added a References page to my advocacy website. It's something I've had in mind to do for a very long time and is a work in progress. It serves as a collection of documentation relating to #CrisisTextLine, all in one place.
https://reformcrisistextline.com/references/
#CrisisLines

#CrisisTextLine is hiring to go "well beyond SMS" to the next gen "artificial intelligence" "natural language processing". Be aware of their spin-off for-profit #LorisAI, which claims sentiment detection & real-time conv. evals, prompts for call centers.

"Crisis Text Line's software engineers develop the next-generation data platform that changes how millions of humans connect with volunteer crisis counselors, our clinical staff and the interaction with one another. Our platform needs to handle information at a massive scale and extend well beyond SMS as a channel. We're looking for engineers who bring fresh ideas from all areas, including information processing, distributed computing, large-scale system design, networking and data storage, security, artificial intelligence, natural language processing, UI design, and mobile."

link to job announce (archived):
http://web.archive.org/web/20231005144819/https://www.crisistextline.org/join-our-staff/
#CrisisLines '#AI ' #NLP

4/ In a 2022 report Crisis Text Line counted about 30,000 children age 10 or younger filled out surveys each year during 2019-2021. About 65,000 children age 11 to 13 completed surveys each year during the same time period. Just back calculate all the conversations sent to their for-profit spinoff Loris .ai back to 2013. Lots of kids. See p41: https://www.crisistextline.org/wp-content/uploads/2022/10/Full-Report-2022_Youth-Resilience.pdf

5/ In the June 2023 order, the #FTC says Personal Info includes “persistent identifiers”. “A persistent identifier that can be used to recognize a user over time & across different websites or online services.” (order, p6)

6/ #CrisisTextLine keeps track of prior callers with a unique identifier tied to their phone #, & links their past transcripts in on the volunteer dashboard. I once spoke with someone with over 200 past conversations. I gave the person the same respect & time I would anyone else.

(cont'd)

4/ In a 2022 report Crisis Text Line counted about 30,000 children age 10 or younger filled out surveys each year during 2019-2021. About 65,000 children age 11 to 13 completed surveys each year during the same time period. Just back calculate all the conversations sent to their for-profit spinoff Loris .ai back to 2013. Lots of kids. See p41: https://www.crisistextline.org/wp-content/uploads/2022/10/Full-Report-2022_Youth-Resilience.pdf

5/ In the June 2023 order, the #FTC says Personal Info includes “persistent identifiers”. “A persistent identifier that can be used to recognize a user over time & across different websites or online services.” (order, p6)

6/ #CrisisTextLine keeps track of prior callers with a unique identifier tied to their phone #, & links their past transcripts in on the volunteer dashboard. I once spoke with someone with over 200 past conversations. I gave the person the same respect & time I would anyone else.

(cont'd)

1/ On June 28, 2023 the #FTC issued a stipulated final order in their case against education surveillance company Edmodo.
Will #CrisisTextLine be next?
https://www.ftc.gov/system/files/ftc_gov/pdf/Edmodo-Dkt15%28Order%20Signed%20by%20the%20Court%29.pdf

2/ From the order: “’Affected Work Product’ means any models or algorithms developed in whole or in part using Personal Information Collected from Children through the Edmodo Platform without Verifiable Parental Consent or School Authorization.”

3/Though the case focuses on Personal Info, I don’t think #CrisisTextLine’s claims of “anonymizing” conversations will protect them. There's the problem of consent during crisis, generally, & in particular, children.
(cont'd)

1/ On June 28, 2023 the #FTC issued a stipulated final order in their case against education surveillance company Edmodo.
Will #CrisisTextLine be next?
https://www.ftc.gov/system/files/ftc_gov/pdf/Edmodo-Dkt15%28Order%20Signed%20by%20the%20Court%29.pdf

2/ From the order: “’Affected Work Product’ means any models or algorithms developed in whole or in part using Personal Information Collected from Children through the Edmodo Platform without Verifiable Parental Consent or School Authorization.”

3/Though the case focuses on Personal Info, I don’t think #CrisisTextLine’s claims of “anonymizing” conversations will protect them. There's the problem of consent during crisis, generally, & in particular, children.
(cont'd)

There's no need to exploit people who reach out to crisis lines for their words, and the words of the workers who answer.

No one who calls those conversations a "massive juicy dataset" should have ever been allowed anywhere near those sensitive, painful, intimate conversations.

Ref link: from October 8, 2015, when Loris .ai was just a twinkle in #CrisisTextLine 's eye.
https://knightfoundation.org/articles/crisis-text-line-expanding-public-access-its-services-and-data-sets/

#CrisisLines #Bioinformatics How is it possible that nonprofits, tech companies & institutions claim to have consent for keeping & using data, when it originates from children under 13, young people 13-17, adults under guardianship...while the person is emotionally distraught?

This is not a trick question. They don't have consent. There is no public relations campaign that brings consent under these circumstances. De-identification is not consent.
And now we can call it what it is: exploitation.
#CrisisTextLine #TheTrevorProject #988Lifeline

📣 New blog post up at my website sadly titled "The Post-Human Movement Away from Human Care"

https://reformcrisistextline.com/the-post-human-movement-away-from-human-care/

ty to: @CriticalAI @emilymbender @lizweil @lilianedwards @timnitGebru @mmitchell_ai
@schock

#CrisisLines #CrisisTextLine #TheTrevorProject #Consent #BioInformatics #Surveillance #NLP #LLM " #AI " #BigTech #MentalHealth #HealthCare #988Lifeline

Good article, difficult subject. I can be picky because of how close I am to this, and clarify:

#CrisisTextLine didn't "tell" #LorisAI to delete the data, it asked. Imagine-the for-profit it created & Crisis Text Line insulated it so well, it couldn't require anything, it "couldn't" dissolve it--which was my request to them when I was still a volunteer. Yes I have all the receipts but so far little influence.

Which is why I'm involved in early beginning stages of a movement to organize #CrisisLines volunteers.

https://www.businessinsider.com/online-therapy-mental-health-apps-betterhelp-talkspace-cerebral-dark-side-2023-4

After pursuing this federal US SAMHSA FOIA (Substance Abuse and Mental Health Services Administration, Freedom of Information Act) request for 8 months, I filed an appeal for lack of response.

If you have nothing better to do, read & weep. Prepare to be discouraged, shameful lack of transparency & accountability for #988Lifeline #CrisisLines, links follow:

Links to readable pdfs:
Appeal:
https://reformcrisistextline.com/wp-content/uploads/Notice_of_Appeal_FOIA-2023-04-17_redacted.pdf

Attachment 1 (my request):
https://reformcrisistextline.com/wp-content/uploads/Notice_of_Appeal_Attachment_1.pdf

Attachment 2 (official response):
https://reformcrisistextline.com/wp-content/uploads/Notice_of_Appeal_Attachment_2.pdf

Attachment 3 (woeful email corresp):
https://reformcrisistextline.com/wp-content/uploads/Notice_of_Appeal_Attachment_3.pdf

#VibrantEmotionalHealth
#CrisisTextLine

If you trust that #SAMHSA and its non-profit grantee #VibrantEmotionalHealth (administering USA #988Lifeline national crisis line system) are caring for people in crisis, you should see them stiff-arm me on legitimate requests for info about the service & private non-profit #CrisisTextLine. Shameful, discouraging, sad. Woeful systems of "care" in USA. Will make public when I can.

This was an #FOIA to SAMHSA and "pretty please" requests to private corporation Vibrant ($48M USD in gov't contract for 988 in 2021, $103M total gov't contracts 2022). Private corps not subject to Freedom of Information Act = accountability shield. What a lack of accountability in a system that is supposed to be serving people at their most vulnerable and powerless moments. I'm not giving up!

I believe there is going to be (and already is) intense pressure for access to crisis conversations as data. Rationalized by de-identifying. This will be justified as helping people in crisis. Driving force will be monetizing & training models for #Bioinformatics, so-called #AI, #HealthBots #MentalHealth screening #Suicide #Depression screening, #Surveillance.

My advocacy is for #Consent and #RespectForPersons. These conversations should be off-limits except for human quality control. Then delete them after say 6 months.

Even research on such sensitive conversations should have consent, including from volunteers. Not possible in crisis moments.

Non-profit corporations in US already have access to the data in-house. Largely unregulated. #CrisisTextLine #TheTrevorProject #VibrantEmotionalHealth (administrator of #988Lifeline).

To overcome the powerful attraction for the data, will require organizing & resistance. On the horizon. Thanks for listening.

The idea to organize crisis line volunteers is not just for numbers, but for networking & communications. This will reveal & make available vast resources of the volunteers & those who support them. The creative collective, the people, the only necessary asset of this organizing. This work is beginning.
#CrisisLines #CrisisTextLine #TheTrevorProject

Three common ethics fails in mental health "AI" "sentiment"/diagnosis research.

1)That you're even doing this.
2)De-identify & waive consent (that's not consent).
3)The use of whatever product you come up with on other persons who will be falsely identified, misled, marketed,etc.

In USA, research is likely headed to product development and monetizing.

Advocacy, awareness are needed to protect sensitive #CrisisLine conversations from being used. They are likely being used now by larger non-profits for in-house research. #CrisisTextLine #TheTrevorProject

https://research.crisistextline.org/what-we-learned-from-training-a-machine-learning-model-to-detect-suicidal-risk-2c65f1d4d9eb

https://www.fastcompany.com/90546991/how-the-trevor-project-is-using-ai-to-prevent-lgbtq-suicides

Three common ethics fails in mental health "AI" "sentiment"/diagnosis research.

1)That you're even doing this.
2)De-identify & waive consent (that's not consent).
3)The use of whatever product you come up with on other persons who will be falsely identified, misled, marketed,etc.

In USA, research is likely headed to product development and monetizing.

Advocacy, awareness are needed to protect sensitive #CrisisLine conversations from being used. They are likely being used now by larger non-profits for in-house research. #CrisisTextLine #TheTrevorProject

https://research.crisistextline.org/what-we-learned-from-training-a-machine-learning-model-to-detect-suicidal-risk-2c65f1d4d9eb

https://www.fastcompany.com/90546991/how-the-trevor-project-is-using-ai-to-prevent-lgbtq-suicides

One year ago today...

#CrisisTextLine announced it would stop sharing data (de-identified crisis conversation transcripts) with its for-profit spinoff Loris .ai, Inc.

This did nothing. Loris had already launched its customer service software. CTL is a shareholder & has a profit-sharing agreement. Now Loris can get its own data from Freshly & MintMobile customers.

But it was a brilliant PR move by Crisis Text Line. All the outrage subsided.

The spinoff of Loris was sophisticated. It was insulated, contractually, and it worked.

Analogy: A thief was caught, allowed to keep all stolen goods, bank them and continue earning income off them.

A reason my advocacy continues.

Crisis Text Line's announcement from a year ago:
https://web.archive.org/web/20220201222519/https://www.crisistextline.org/blog/2022/01/31/an-update-on-data-privacy-our-community-and-our-service/

@gvwilson @seresearchers

B. See if this poem helps with understanding. I'm a former volunteer with #CrisisTextLine
https://reformcrisistextline.com/some-people-want-our-words/

C. Follow and support these individuals, and others doing this work. They are so much more knowledgeable than me.
@Mer__edith
@safiyanoble
@emilymbender
@timnitGebru
@kendraserra

Thanks again to you, for your care and your place in the system to make a real difference.

I submitted these comments about #dataethics #consent #privacy #AI to the USA Federal Trade Commission, using #CrisisTextLine for examples. Also #TheTrevorProject is mentioned.
https://reformcrisistextline.com/wp-content/uploads/2022/11/FTC_Comment_tdr_2022-11-21.pdf

Link is to a blog post with instructions for deleting conversation data from Crisis Text Line. For your use or for sharing with persons who may use the service.
#CrisisTextLine #CrisisLines
https://reformcrisistextline.com/how-to-delete-data-from-crisis-text-line/