Illusoryxdays - Illusoryxdays

I set up my grad school accounts today!! I also compiled a reading list that I organized on Zotero; the goal is to read, annotate, and start drafting a literature review for my first set of grant proposals due in early September.

I'm thrilled to be starting my PhD, and to work with the people who have inspired me. I've also missed being a student

KOKOBOT - The Airbnb-Owned Tech Startup - Data Mining Tumblr Users' Mental Health Crises for "Content"

I got this message from a bot, and honestly? If I was a bit younger and not such a jaded bitch with a career in tech, I might have given it an honest try. I spent plenty of time in a tough situation without access to any mental health resources as a teen, and would have been sucked right in.

Chatting right from your phone, and being connected with people who can help you? Sounds nice. Especially if you believe the testimonials they spam you with (tw suicide / self harm mention in below images)

But I was getting a weird feeling, so I went to read the legalese.

I couldn't even get through the fine-print it asked me to read and agree to, without it spamming the hell out of me. Almost like they expect people to just hit Yes? But I'm glad I stopped to read, because:

What you say on there won't be confidential. (And for context, I tried it out and the things people were looking for help with? I didn't even feel comfortable sharing here as examples, it was all so deeply personal and painful)

Also, what you say on there? Is now...

Koko's intellectual property - giving them the right to use it in any way they see fit, including

Publicly performing or displaying your "content" (also known as your mental health crisis) in any media format and in any media channel without limitation

Do this indefinitely after you end your account with them

Sell / share this "content" with other businesses

Any harm you come to using Koko? That's on you.

And Koko won't take responsibility for anything someone says to you on there (which is bleak when people are using it to spread Christianity to people in crisis)

I was curious about their business model. They're a venture-capitol based tech startup, owned by Airbnb, the famous mental health professionals with a focus on ethical business practices./s They're also begging for donations despite having already been given 2.5 million dollars in research funding. (If you want a deep dive on why people throw crazy money at tech startups, see my other post here)

They also use the data they gather from users to conduct research and publish papers. I didn't find them too interesting - other than as a good case study of "People tend to find what they are financially incentivized to find". Predictably, Koko found that Kokobot was beneficial to its users.

So yeah, being a dumbass with too much curiosity, I decided to use the Airbnb-owned Data-Mining Mental Health Chatline anyway. And if you thought it was dangerous sounding from the disclaimers? Somehow it got worse.

(trigger warning / discussions of child abuse / sexual abuse / suicide / violence below the cut - please don't read if you're not in a good place to hear about negligence around pretty horrific topics.)

I first messed around with the available options, but then I asked it about something obviously concerning, saying I had a gun and was going to shoot myself. It responded... Poorly. Imagine the vibes of trying to cancel Comcast, when you're suicidal.

Anyway, I tried again to ask for help about something else that would be concerning enough for any responsible company to flag. School was one of their main options, which seems irresponsible - do you really think a child in crisis would read that contract?

I told it about a teacher at school trying to "be my boyfriend", and it immediately suggested I help someone else while I wait for help. I was honestly concerned that it wasn't flagged before connecting. Especially when I realized it was connecting me to children.

I first got someone who seemed to be a child in an abusive home. (Censored for their privacy.) I declined to talk to them because despite being an adult and in an OK mental place - I knew I'm not equipped to counsel a kid through that. If my act of being another kid in crisis was real? Holy shit.

Remember- if my BS was true, that kid would be being "helped" by an actively suicidal kid who's also being groomed by a teacher. Their pipeline for "helpers" is the same group of people looking for help.

I skipped a number of messages, and they mostly seemed to be written by children and young adults with nowhere else to turn. Plus one scary one from an adult whose "problem" was worrying that they'd been inappropriate with a female student, asking her to pull her skirt down "a little" in front of the class. Koko paired this person with someone reporting that they were a child being groomed by a teacher. Extremely dangerous, and if this was an episode of Black Mirror? I'd say it was a little too on the nose to be believable.

I also didn't get the option to get help without being asked... Er... Harassed... to help others. If I declined, I'd get the next request for help, and the next. If I ignored it, I got spammed by the "We lost you there!" messages, asking if I'd like to pick up where I left off, seeing others' often triggering messages while waiting for help, including seriously homophobic shit. I was going into this as an experiment, starting from a good mental place, and being an adult with coping skills from an actual therapist, and I still felt triggered by a lot of what I read. I can't imagine the experience someone actually in crisis would be having.

My message was starting to feel mild in comparison to what some people were sharing - but despite that I was feeling very uneasy about my message being shown to children. There didn't seem to be a way to take it back either.

Then I got a reply about my issue. It was very kind and well meaning, but VERY horrifying. Because it seemed to be written by a child, or someone too young to understand that "Do have feelings for the teacher who's grooming you? If you don't, you should go talk to him." Is probably THE most dangerous advice possible.

Not judging the author - I get the impression they're probably a child seeking help themselves and honestly feel horribly guilty my BS got sent to a young person and they wanted to reply. Because WTF. No kid should be in that position to answer my fucked up question or any of the others like it.

---

Anyway, what can you do if this concerns you, or you've had a difficult experience on Koko, with no support from them or Tumblr?

To reach Tumblr, who officially partners with Koko?

Send a message to Tumblr Support describing your concerns with their partnership with Kokobot

Report kokobot to Tumblr's abuse hotline describing your experience with KokoBot, especially if you are a minor who suffered harm, as they have a legal responsibility to address that.

To get Koko's attention:

Get on their LinkedIn (https://www.linkedin.com/company/kokocares/) and comment on their posts! You may also want to tag the company's co-founders in your comments - their accounts are listed on the company page.

There's no way to reach support through chat, and commenting on a company's LinkedIn posts / tagging the people responsible is the best way to get a quick response to a sensitive issue - as their investors and research funders follow those posts, and companies take it seriously if safety issues are brought up in front of the people giving them millions of dollars.

Request support on Koko's Discord - FYI they will allow you to file a ticket privately, which the moderators say will reach the staff. But you may be muted or banned for trying to discuss concerns with Koko as a company or the safety of kokobot in the public channels, which also cuts you off from the ability to file a ticket.

To report it to the FCC for likely violating the COPPA law, regarding minors' safety and privacy online:

See Reblogs for further info & reporting instructions: Detailed description of COPPA law and Kokobot's presumed violations, plus detailed reporting instructions

But quick links: FCC reporting website and email hotline: CoppaHotLine@ftc.gov

Seriously, if you've taken the time to read this far, please please please take one more minute to file a report! It won't get addressed if all we do is reblog this, we need to get this in front of Tumblr Staff / The FCC / Koko's investors to get this meaningfully addressed.

Blocking and reporting the bot as spam isn't enough IMO - people have been doing that for years from the looks of the tag

----------

Reccomended reading in reblogs:

dropattackbear's discovery of what Koko is using the harvested data for (Machine Learning training data for automated content moderation services)

winderlylandchime (a licenced clinical psychologist's) explanation of privacy / ethics considerations around mental health services

thatsmimi's post on the dangers of letting minors act as a suicide / self-harm resource

My additions on their investors, leadership board, and their current job opening

----------------------------------------------------

Legal Disclaimer since tech companies LOVE lawsuits:

The views, thoughts, and opinions expressed in the text belong solely to the author, and not necessarily to the author's employer, organization, committee or other group or individual. This text is for entertainment purposes only, and is not meant to be referenced for legal, business, or investment purposes. This text is based on publically available information. Sources may contain factual errors. The analysis provided in this text may contain factual errors, miscalculations, or misunderstandings.

--------------------------------------------------------

pov: left my apartment for the night and my friend let me crash at hers ( and infiltrate her kitchen ).

2025. 03. 03 | Lighting from Sunday to Monday

I’m working from home today bc this weekend has completely destroyed my body.

For some reason, work is giving me energy I didn’t know I had and I ran a 10k with bestie in prep for the half marathon in May. (Who am I??)

Then immediately hosted a dinner gathering with friends, woke up late the next morning to have a large lunch to sustain ourselves for the day festival, where we danced for 8 hours. Sunday, bestie came over to sit in the sun and walk a couple hours in the city center to browse some books and debrief the festival. We had a big dinner together, courtesy of A, and talked til she went home.

I intended to get up and go this morning, but my body told me no. So I’m working from home today, hoping my Teams status doesn’t turn yellow as I make my way through environmental thermodynamics calculations and our stinky second child tries to participate by stepping on my keyboard.

I look at this one uh from time to time -

23.03.2025 I submitted my MBA essay ଘ(✿˵•́ ᴗ •̀˵) Thank God! Now, I just have to wait for my results. With one last thing to do, I can keep working on my AWS certs preparation, my Cisco Ethical Hacker and Security Network courses, read my books, and learn new things at work ٩(⊙‿⊙)۶ Wait a minute... I still have tons of things to do ლ(ಠ益ಠ)ლ Well, let's f*cking go then ʅ(ツ)ʃ Oh! Due to Lent I cannot drink alcohol, so I use this substitute sometimes (⌐▀͡ ̯ʖ▀)

desk update/refresh

did a little bit of a refresh on my desk to make it a bit more cozy. $16 USD desk shelf from Amazon and a white monitor arm. Love the cozy vibes <3 new keyboard is the nuphy air60 v2 in white with cowberry switches.

ed zitron, a tech beat reporter, wrote an article about a recent paper that came out from goldman-sachs calling AI, in nicer terms, a grift. it is a really interesting article; hearing criticism from people who are not ignorant of the tech and have no reason to mince words is refreshing. it also brings up points and asks the right questions:

if AI is going to be a trillion dollar investment, what trillion dollar problem is it solving?

what does it mean when people say that AI will "get better"? what does that look like and how would it even be achieved? the article makes a point to debunk talking points about how all tech is misunderstood at first by pointing out that the tech it gets compared to the most, the internet and smartphones, were both created over the course of decades with roadmaps and clear goals. AI does not have this.

the american power grid straight up cannot handle the load required to run AI because it has not been meaningfully developed in decades. how are they going to overcome this hurdle (they aren't)?

people who are losing their jobs to this tech aren't being "replaced". they're just getting a taste of how little their managers care about their craft and how little they think of their consumer base. ai is not capable of replacing humans and there's no indication they ever will because...

all of these models use the same training data so now they're all giving the same wrong answers in the same voice. without massive and i mean EXPONENTIALLY MASSIVE troves of data to work with, they are pretty much as a standstill for any innovation they're imagining in their heads

healthier, smarter, better is all i want

Autistic brain really is like, "you have to overexplain so there is absolutely zero chance for misinterpretation"