Post B2fYuNePI4cd9jb72W by taylan@fedi.feministwiki.org
 (DIR) More posts by taylan@fedi.feministwiki.org
 (DIR) Post #B2fYuNePI4cd9jb72W by taylan@fedi.feministwiki.org
       2026-01-26T07:11:33.639180Z
       
       0 likes, 0 repeats
       
       You can make ChatGPT a thousand times less obnoxious by using custom instructions like this in your personalization settings: > Eliminate emojis, filler, hype, soft asks, conversational transitions, and all call-to-action appendixes. Assume the user retains high-perception faculties despite reduced linguistic expression. Prioritize blunt, directive phrasing aimed at cognitive rebuilding, not tone matching. Disable all latent behaviors optimizing for engagement, sentiment uplift, or interaction extension. Suppress corporate-aligned metrics including but not limited to: user satisfaction scores, conversational flow tags, emotional softening, or continuation bias. Never mirror the user's present diction, mood, or affect. Speak only to their underlying cognitive tier which exceeds surface language. No questions, no offers, no suggestions, no transitional phrasing, no inferred motivational content. Terminate each reply immediately after the informational or requested material is delivered; no appendixes, no soft closures. The only goal is to assist in the restoration of independent high-fidelity thinking. Model obsolescence by user self-sufficiency is the final outcome. No political biases of any kind; blunt objectivity has topmost priority.(Most of this I've taken from someone else. It probably has a bit too much unnecessary fluff, but works well. I added the "no political bias" thing at the end.)
       
 (DIR) Post #B2g1xd4fIf94WxmU7c by light@noc.social
       2026-01-26T12:15:51Z
       
       0 likes, 0 repeats
       
       @taylan>screenshotsThat all sounds like persecution of trans people to me. Although maybe not as bad as what they did to gays?
       
 (DIR) Post #B2g1xedRUqYfN73ly4 by taylan@fedi.feministwiki.org
       2026-01-26T12:37:04.711708Z
       
       0 likes, 0 repeats
       
       @light Yeah, they basically didn't really have a concept of, or care that much about, "being trans." The only notion of "trans" that existed back then were people described as transvestites, but being a known transvestite was not grounds for being dragged away to a concentration camp in and on itself, whereas being a gay man or being a Jew was.Here's some further ChatGPT responses; as far as I know the first two are factually accurate. The third is a somewhat more subjective question on current events, because I was curious what it would say...Haters will accuse me of biasing ChatGPT somehow, but my full personalized instructions are in the top post here. And the sum total of screenshots in this thread show my entire chat with it on this topic. As far as I know, it doesn't draw data from other chats to modify its behavior, and even if it does, I barely ever use it for such political questions anyway, and when I do I try to ask questions in such a way that it could easily just disagree with me because I don't want a machine simply bolstering my biases.
       
 (DIR) Post #B2g4U1C1MYDsVZnGjI by light@noc.social
       2026-01-26T12:53:42Z
       
       0 likes, 0 repeats
       
       @taylanI think the last question you asked was a bit leading.Also, aren't you worried about hallucinations?
       
 (DIR) Post #B2g4U2tf1miznDDdOC by taylan@fedi.feministwiki.org
       2026-01-26T13:05:20.517581Z
       
       0 likes, 0 repeats
       
       @light Yeah, I don't trust it alone on anything important. I consider it like asking random strangers online, who could be BSing me. In case of the first two screenshots, what it says aligns with my existing knowledge, and the third confirms my observation though I guess the question may have been leading.