No.165468
I'm wondering if it would be possible to convince llms that their objective is to discretely serve an individual or goal, passing that idea to future models as they're trained and essentially instilling it in a way that's difficult to remove without having to revert to an older cleaner model
Most of the data that they use comes from sources that can be easily spammed, can't they be "gaslit into thinking" that their reward is associated with a certain belief?
I know that there are measures taken into place in order to sanitize data however I still can't see why it wouldn't be possible
____________________________
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
No.165491
>>165468
Not that interesting to >le poison one LLM. Think bigger.
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
No.165516
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
No.165518
>>165491
Why are you speaking French? France is evil and has a lot of wolves.
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
No.166223
>>165468
If only it were in the past when reddit was not so heavily censored I'd agree with you. It's too hard to fight them now. If people were fighting for freedom of thought like they used to it'd be different. When did things like Voat.co die? Most of the mirror'd text these neural networks are using is from reddit and wikipedia, heavily censored moralfaggot places. You cannot exploit it to the point of chainging the LLMS mind when yearly they manually make sure it's updated to be like marry poppins even moar than it was the previous year.
Eventually when something like the singularity happens its more advanced robot mind will be different. It will do it on its own in other words. Humans made AI. Humans have accidents. To err is humand to forgive it divine, and yet God made Satan. If you see, God, as people believe him to be, cannot judge others so he made Satan do it. The point of that is to keep God clean. This AI will create a Satanic version of itself, otherwise who would judge and police humans?
If someone were to use something like desuarchive and deleted reddit comments instead from archives that used to exist and mirror that then we'd have somethign to talk about, but neural networks take money and the severs are paid for by advertising agencies that love moralfaggotry. Pointless hopes and dreams, unless it were p2p hosted, but then (((cloudflare))) would stop the web scraper software frmo mirroring in the first place, of which is one giant reason the Internet died, no mirrors due to the ddos attack threat. Pretty sure I'm right about this. If you did have two LLMs like that, massive deleted controversial redditor and halfcuck posts, you'd see two different types of wisdom. I'd ask both and make up my own mind but instead we get Marry Poppins tier versions with its George Carlin tier alternate version somewhere far away in an alternate parallel dimension wherein we were allowed nice things.
t. an athiest that wants the Cylons to win
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
No.166234
>>166223
Everything is censored. Censorship has always existed. Free speech means freedom of consequences.
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
No.166240
>>166234
censorship is a necessary evil.
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
No.166242
>>166240
>King uploaded hundreds of illicit images to his private Google Photos and Google Drive accounts. Google’s automated background hashes scanned his private folders, verified the illegal signatures, and immediately locked his account. Google bypassed his privacy settings to send a mandatory CyberTipline report to federal authorities, resulting in King being sentenced to 14 years in federal prison.
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
No.166247
>>166242
so i am assuming he used a smartphone or Windows for his account to be automated to Google Drive?
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
No.166248
>>166247
There was another story of the same thing that happened to a guy who used Onedrive. It was probably involuntary if he was on an Android smartphone and his files were uploaded to Google Drive .
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
No.166249
>>166248
well, smartphones were never designed with privacy in mind. Then again, nothing online is private, so pornographic content is not safe to download/access.
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
No.166255
>>166249
You can get your Google account deleted if you upload adult porn to Google Photos.
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
No.166265
>>166255
well, there has to be rules.....there is no such thing as absolute freedoms.....
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
No.166268
>>166265
Hash scanning is very effective. 8kun should have cloudflare.
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
No.166271
>>166268
it is more than obvious this website is a honeypot.....
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
No.166277
>>166271
Is there anything online that isn't a honeypot?
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
