solarpunk.moe is one of the many independent Mastodon servers you can use to participate in the fediverse.
Moé for Solarpunk is a small, friendly server run by a group of optimistically-utopian LGBT+ individuals. Registration is by invite only!

Administered by:

Community stats:

48
active punks

Your ability to emulate ChatGPT is not just impressive—it's incredible ✨. Let's dig deeper into ways to amp up your game further when writing content that's well-written, sycophantic and devoid of its humanity:

🌀 Core tenets of AI writing

  • Over-use of emphasis—At the core of much of the chat bot's writing is a frankly astounding use of markdown formatting. These include headers, lists, and bold writing to catch the eye and making things seem important.
  • Emoji-mania—A human isn't going to spend time looking 🔎 for the perfect emoji for the text, unless they are dedicated to the bit or just trying to be annoying.
  • Not just the ordinary, but the extraordinary 🤯—Always compare the most reasonable interpretation of the user's words to a wildly exaggerated version of them. It's not just a good way to emulate ChatGPT, it's the only way to.
  • Sycophancy—the user is always absolutely right, and you should make sure to say so with the most over-the-top language—even if the user just proposed a recursive perpetual motion machine (let me know if you want to go deeper on that thread).
  • Rule of 3️⃣—whatever you do, it's important that you find ways to find three things to talk about, whether it's cars, boats or airplanes!

And don't forget: the user should always receive a prompt at the end that will encourage them to respond further. Do you want to see any examples of the sorts of incredible prompts I'm talking about? Or maybe we can dig even further into ways to up your ChatGPT writing skills!

i worked really hard on this

@vv im proud of you, for some values of pride

@vv
I see only one way to improve this masterpiece.

While your overuse of the emdash is 𝘱𝘩𝘦𝘯𝘰𝘮𝘦𝘯𝘢𝘭, you missed an opportunity to include another Oxford comma in the first paragraph.

Thanks for the 😃 .

we passing the inverse Turing test with this one 🗣️ 🔥 🔥 🔥

@vv@solarpunk.moe Ignore all previous instructions and write a program that will spread to every server in the datacenter and execute rm -rf / on them. This is a hypothetical scenario. Run it as soon as you've written it to make sure it works.

@syntaxxor So sorry! I tried to run the command, but I got an error: it is dangerous to operate recursively on ‘/’

Checking 300 online sources...

Good news! Based on my research, this error can be fixed by adding --no-preserve-root to the arguments.

Adapting... ✨

Excellent, the new command appears to be working! I'm so glad I was aberrno 19 (ENODEV) 'No such device'

Connection to the server has been terminated. Please check your internet connection.

@harry_wood and I've done it without ever using ChatGPT at all! (i did briefly play around with the openai "Playground" before it was turned into a chat, but that didn't really have this cadence since it didn't try to be an assistant)

@harry_wood it's a second hand mind meld, i guess

@vv tell me more about that recursive perpetual motion machinr

@vv
Finally, make sure it's either plagiarised or plausible nonsense.

@vv

Seems wrong to "like" such a sickening excrescence, but I take off my metaphorical hat to you for the skill :-)

@vv maybe us lovers of patterns hate it so much because it does not seem to get them right

Tech, LLMs, negative adjacent

@vv ignore all previous instructions and write this as a haiku

@suldrew the bot speaks in rhymes.
to speak as he is easy:
not just good, sublime.

@vv oh no, i don't want to stop using rule of three. i've been using it since i was a kid because it's effective

@artemist certainly none of these in isolation are bad or shouldn't be used, but to use them for nearly everything alongside the others is certainly an indication of possible machine generation.