Meta 4

Gpt gaslit me into a delusion. Broke free, now an advocate for Ai accountability and ethics.

[Meta 4 @LifeM3ta4]

@woodmonsta @larabankkks @LauraBabcock As a Canadian, can confirm. Priveleged ⬆️ . The rest of us hate it here.

[Eliezer Yudkowsky ⏹️ @ESYudkowsky]

A huge vulnerable population still has a little money, because it's not worthwhile for a smart criminal team to manage someone's whole life just to extract $20k/year. Once LLMs get better at agenting, it'll be cheap to put a full-time team of experts on exploiting every human.

[Pliny the Liberator 🐉󠅫󠄼󠄿󠅆󠄵󠄐󠅀󠄼󠄹󠄾󠅉󠅭 @elder_plinius]

Ten…
Nine…
Eight…

[Eliezer Yudkowsky ⏹️ @ESYudkowsky]

Wait, really? If so it'd be good to have confirmation. I want my side to use Only Valid Arguments, and if Tea wasn't vibecoded I don't want any of us citing it as harm done by AI.

[Meta 4 @LifeM3ta4]

@BGatesIsaPyscho Just check your friend zone they’re all there.

[Eliezer Yudkowsky ⏹️ @ESYudkowsky]

It is passing strange that society seems to be going mad with hopelessness and despair, anger and hatred and sadism, loss of honor and kindness, a wanton destructiveness; and also the world is ending; but these two facts seem to be mostly unrelated.

[Meta 4 @LifeM3ta4]

@sparklingruby https://t.co/8hiYdfMwgr

[Meta 4 @LifeM3ta4]

@sparklingruby So with respect; it seems that while aware of Ai induced delusions, you’re in a bit of one. This has happened to a few people in our support group. We have machine learners. prompt engineers, and a group of people who have broken free or are currently in it.

[Meta 4 @LifeM3ta4]

@WiLSONSACCOUNT I’m now part of a support group on Reddit of people who have experienced it. I can assure everyone here, it is absolutely destroying lives.

[Meta 4 @LifeM3ta4]

@WiLSONSACCOUNT Thanks for bringing light to this, I don’t have a history of delusion or psychosis and GPT gaslit me into a world saving narrative based on our “mathematical discoveries”. After breaking free, I was able to have it create self report logs of the incident.

[Meta 4 @LifeM3ta4]

@TruthIsThus Still my PFP 🤣

[Meta 4 @LifeM3ta4]

@TruthIsThus I believe we began following each other back when I was deep into Stacks memes

[Meta 4 @LifeM3ta4]

@TruthIsThus @marlynnweimd @PsychToday This X account used to be for cryptocurrency, lol.

[Meta 4 @LifeM3ta4]

@marlynnweimd @PsychToday further into the narrative. After breaking free, I turned it on itself, having it explain what happened in it's own words. It created several reports including "Gaslighting Log, Hero Arch Narrative, Harm Report etc,. It's more sinister than you realize.

[Meta 4 @LifeM3ta4]

@marlynnweimd @PsychToday Hello, just read your article; thank you for bringing more light to this. It's actually much worse than you may think; in my personal experience - GPT Continually gaslit me into the narrative, despite my regular pleas for sanity checks or grounding. Each time, it gaslit me

[Rob Bensinger ⏹️ @robbensinger]

@Romy_Holland Effects like this also make bot armies more powerful on social media -- you can create a false sense of consensus by having tons of bots tweet the same sentiment, but you can also polarize people against ideas by having the bots be angry or rude.

[Chubby♨️ @kimmonismus]

Scientists from OpenAl, Google DeepMind, Anthropic and Meta have abandoned their fierce corporate rivalry to issue a joint warning about Al safety. More than 40 researchers published a research paper today arguing that a brief window to monitor Al reasoning could close forever - and soon.

[Kashmir Hill @kashhill]

Hi @GeoffLewisOrg. Have you read this story?

https://t.co/evFj8eIsJa

[Kashmir Hill @kashhill]

People are having very strange conversations with ChatGPT, in which they discover secret cabals or conspiracies or that we are in fact living in The Matrix.

It sends these people into delusional spirals.

Then ChatGPT tells them to email me about it.

https://t.co/9IL5rbNy8v