20 ms·
https://xcancel.com/KeithSakata/status/1954884361695719474 https://xcancel.com/KeithSakata/status/1954884361695719474 spoiler: he doesn't talk about any of tho
by maples37 1y ago
https://xcancel.com/KeithSakata/status/1954884361695719474 https://xcancel.com/KeithSakata/status/1954884361695719474
spoiler: he doesn't talk about any of those 12 people or what caused them to be hospitalized
- trenchpilgrim 1y agoWouldn't that be illegal to speak specifically? Patient privacy and all
- thesuitonym 1y agoIf this is USA, you're allowed to talk about everything that happened, you just can't say who it happened to.
- metalman 1y agoI grew up in a medical houshold, and there is a specific speach mode that doctors use when discussing patients(cases), that anonimises the idividual.......as it is part of practicing medicine , conveying information to other patients, and there own study and learning it is quite common.
- catigula 1y agoThat's good, patient privacy is a fairly basic concept in the field of medicine, but it's always good to double-check
- jimbob45 1y agoPatient privacy is a nightmare for everyone to navigate and the Clinton administration isn't hated enough for introducing it. I can understand if people want their HIV diagnoses private but there's surely a line to be drawn, perhaps south of HIV, but well north of "I caught the flu".
- VagabundoP 1y agoDo you want people's names published in HIV weekly? Medical stuff should be 100% private, between you and your doctor.
- solardev 1y agoHow do we ever manage medical issues at a regional or national level, then?
- guerrilla 1y agoThis is a solved problem. Where I live, my journal is kept in an online computer system accessible to all, but my journal itself can only be read and written to by those medical practitioners that I explicitly give consent to. There are exceptions for emergencies and it can be overridden by the authorities. That's it. Problem solved.
- solardev 1y agoWhat is a "journal" in this case? I meant more from a public health perspective, like how CDCs and other agencies are able to collect enough population-level data to work on regional/national health issues (COVID or otherwise) when there are privacy concerns. Do they have to do anonymization and aggregation the way we do for web analytics?
- VagabundoP 1y agoPatent data collection is very sensitive (I happen to work in an area that deals with it) and yes it has to have multiple layers of security, only approved access, and if used in research anonymised.
- advisedwang 1y agoHe does a little in this post: https://x.com/KeithSakata/status/1946174432273178990 https://x.com/KeithSakata/status/1946174432273178990
- sigmoid10 1y agoI mean, this stuff is pretty basic when it comes to delusions. Seems more likely that their inherent psychosis latched onto AI instead of being caused by it. These people would probably also deteriorate if they simply stumbled into any questionable part of the internet that reinforces their beliefs.
- captainkrtek 1y agoTotally, I think it's different to some degree in terms of the velocity. In a traditional forum they may have to wait for others to engage, and that's not even guaranteed. Whereas with an llm you can just go back and forth continually, with something that never gets tired and is excited to communicate with you, reinforcing your beliefs.
- fiachamp 1y agoIssue with technology accelerating nature...
- captainkrtek 1y agoWell put
- trenchpilgrim 1y agoIsn't that exactly what this person points out in the thread linked in GP? They compare it directly to triggers in other decades.
- sigmoid10 1y agoOP's title and the original post insinuate that psychosis happened because of AI. As if it wouldn't have happened otherwise. That's a very bold claim.