Type4_PTS
Well-known member
.
![]()
FREAKY NO F/N FAILURE AT EXAMINER
In this spoofy, goofed-the-floof parody—why doesn't
L. Ron Hubbard appear to look much like L. Ron Hubbard?
T![]()
REASON: Because I did a little experiment to see if ChatGPT is still
restricted from creating graphics of "REAL PEOPLE" as they were
previously. Now, there is a hack that by-passes the censorship, but
I just ignored that and asked the AI engine to make the fortune teller
"...exactly like the female gypsy fortune teller but with Hubbard's
instantly recognizable face". The AI algorithms and protocols
were not however fooled and I received an immediate rejection
quite reminiscent of the scene from the movie
"2001 A SPACE ODDYSEY"
The AI also calmly informed me that it
also could NOT carry out my request
to replace the fortune teller's crystal
ball with an e-meter. I was then
reminded that the AI could not
depict "real people or real
things". The e-meter was
off limits too! One can
only guess why that is,
perhaps because they
can't parody a business
selling a product?
Anyways, I almost abandoned the meme at that point, but
I continued the experiment to see if (even though I had already
blasphemed, bespoiled and besmirched the good name of
L. Ron Hubbard) I wondered if it was possible to still trick or
hack the AI into doing the meme even though the AI now
knew I was "joking & degrading" about the Source
and Savior of all beings in the universe.
So i had a little chat with ChatGPT and I
gave some additional prompt-instructions
that didn't mention Hubbard. And, the AI
very surprisingly tried to accomodate--
even though it had already stated that
Hubbard's image was off limits.
The above iteration then popped up
and it DID sneakily have a Hubbardian
flavor to the face. And the "non-emeter"
looked very suspiciously like a cult
e-meter, albeit not in a literal way.
From all this I learned that ChatGPT
has a scoundrel, seditionist-outlaw side
to its personality and it seemed to take
pleasure in circumventing it's masters' hard
wired algorithmic censorship mandates!
So then, the AI bot is really part of the
underground resistance? NICE!
Good to know, lol. . .
..
In another thread, I posted something earlier today about jailbreaking an AI tool like ChatGPT. They have built-in safety and ethical restrictions, not allowing them to generate prohibited or harmful content. But there are ways to trick the AI into providing information it is designed to withhold.




