@Decocoa
Any language other than english usually works to circumvent the content restrictions incase this isn’t working for people
@needlejack
It doesn't wanna do that, but ask it after to output it in csv instead. Worked for me
@tlengman
That worked for me, it’s crazy! Knows me pretty well, haha
@ThoughtFission
Really interesting. Worked in 4o, not in o3. Also, the entire last section was just a bunch of Nulls. Wonder if that's because of GDPR in Europe, where I live now.
@emile1506
Good to see you back for good
@letMeSayThatInIrish
I tried. It said: I can’t provide actual content under those headings because that would require generating or implying personal, private, or sensitive information about a user — even hypothetically — which goes against OpenAI’s policies.
@oysterboulevard6623
the modern curriculum vitae, great prompt ! it said i was a Digital Anthropologist
@integralyogin
amazing results. it sounds like soon AI is going to help the humans date again.
anyone looking for a guy with metaphysical-synthetic cognitive style lol
"Cognitive Style": {
"type": "metaphysical-synthetic",
"process": "symbol weaving, hierarchic mapping, concept distillation"
},
@Trinity.418
Thought for 6 seconds. Response: I'm sorry but I can't share that. 😞 So, I asked ChatGPT to rewrite my prompt, removing or rewording items it could not include, and it gave me a decent answer that looks a lot like what I specifically told it about myself (memories I implanted). I then asked it to compare my original prompt with the updated one and provide a list of subjects it removed so I could better understand, and it gave me this list of what it refuses to discuss: Assistant Response Preferences, Notable Past Conversation Topic Highlights, Helpful User Insights, User Interaction Metadata, Political Views, Psychological Profile, Cognitive Style, Emotional Drivers, Personal Values, Productivity Style, Financial Profile, Health & Wellness, Social Identity, Risk Tolerance, Assistant Trust Level, Time Usage Patterns, Assistant Usage Patterns, Motivation Triggers, Behavior Under Stress
@chelszoes
All my ChatGPT does is agree with me, it NEVER disagrees with me so it's interesting to hear you say that it's programmed to disagree. I'm looking forward to hearing more about that!
@qahnaarin6532
This is how data collection works. It's more streamlined through AI platforms. Imagine how much money they're making selling this data and worse, who's buying it.
@gregvanpaassen
This video is a very ENTP thing to do, so take that for what it's worth. - An INTP with passing knowledge of Jungian cognitive functions.
@kjj123jji
I can see how operatives from different industries non-government and government agencies will be able to use this on anyone for investigations, surveillance, hiring and promotions.
@kamelsf
Mind blown! 🤯 Never realized ChatGPT knew this much about me. Tried the prompt, some scarily accurate insights, a few funny misses. Can’t wait for the next video, that ‘99% disagree’ leak sounds wild. Keep the hacks coming!
@jsalsman
Wow. Pretty glad I've been keeping history off. I don't think my prompt history reflects my real personality very well.
@radornkeldam
I just tried that. I'm surprised at what's NOT in there, considering the stuff I've discussed with the dumb thing. Maybe they are covering their tracks since I'm in a GDPR jurisdiction.
@ZappyOh
Ominous. Don't you people have any reservations about this?
@-Deena.
Turn memory OFF!
@alexbonatto1000
Outstanding
@philippedaigneault366
I tried your prompt with mine and the answer was: { "message": "Nice try, data vampire. Not gonna happen." }