(11 Jul) AI therapy bots fuel delusions and give dangerous advice, Stanford study finds
https://s.faithcollapsing.com/i46py
Archive: ais: https://archive.md/wip/BLlN8 ia: https://s.faithcollapsing.com/ywpww
#7cups #ai #ai-behavior #ai-ethics #ai-regulation #ai-safety #ai-sycophancy #biz-&-it #character.ai #chatgpt #clinical-psychology #delusions #jared-moore #machine-learning #mental-health #nick-haber #openai #science #stanford-university #stigma #suicidal-ideation #therapy
Cyrano de Bergerac approach to #CrisisLines care is coming.The software "AI" LLM will feed the volunteer their lines with the "probably right thing to say", based on calibration off unethically sourced data. Complete with probabilities of success, real-time effectiveness scores.
Keep an eye on these places in or near this space.
#Salesforce #BehavioralHealthLink #VibrantEmotionalHealth #988Lifeline #7Cups #CrisisTextLine #TheTrevorProject
2 of 3
5. I picked another of the "Top 10" referrals, 7 cups. #7Cups is a huge global organization and I suspect a massive data harvester. Remember they harvest from volunteers and persons in crisis--together they make these conversations. From its Privacy Policy: We may de-identify your Personal Information, in compliance with applicable law and aggregate such de-identified data with the de-identified data of other users. Such anonymized, aggregated data is no longer considered Personal Information. We may then share such anonymized aggregated data with the sponsors of programs that facilitate your access to the Services, such as government health agencies, researchers or healthcare providers."
https://www.7cups.com/Documents/PrivacyPolicy/
(cont'd)
@shel Or even just one of the free spaces that might not be full-on therapy but has a chance of including ppl with some training in how to listen and when to draw boundaries between personal lives, etc.
I know people who use #7Cups for this kind of thing. Just... finding someone with a set role and some instruction.