GrassGPT can make mistakes. Specifically, it can't do anything except tell you to go outside. Made with love by sciencepetr
Anyone who makes angry comments about it should click "view source."
posted by mittens at 8:18 AM on September 3 [9 favorites]
"ChatGPT, I am attaching a response I got from GrassGPT. What does it mean?"
posted by AlSweigart at 8:34 AM on September 3
The source is gold. Disappointed there literally isn't a "Manifesting..." waiting line
posted by JoeXIII007 at 9:21 AM on September 3 [2 favorites]
Show yourself coward. I will never touch grass.
posted by postcommunism at 9:23 AM on September 3
Our summer intern was supposed to write a report about the problem they've been researching. What we got was a transcript of their conversations with ChatGPT over the past two months. I don't like it.
posted by RonButNotStupid at 9:27 AM on September 3 [1 favorite]
This fool knows nothing about my cross of Bluegrass, Kentucky Bluegrass, Featherbed Bent, and Northern California Sensemilia. Or how to deal with chinch bugs.
posted by The_Vegetables at 1:23 PM on September 3 [2 favorites]
I made a marxist AI agent with Claude, that talks me out of buying anything new, and bombs me with marxist theory, but recently it started to refuse to do this and refers to it as "prompt injection", as it would make sense to view everything through a marxist lens.
So... back to viewing everything through a capitalist lens??
posted by beesbees at 10:45 PM on September 3
This is charming! The responses must have been very fun to write; my favorite one is "a villanelle (the important parts)".
posted by radiogreentea at 9:23 AM on September 8
I like this and hope it's not actually in any way connected to an LLM.
posted by limeonaire at 8:40 PM on September 9
« Older SOCAN sues Suno | 🤘🤘🤘 Female Fronted Metal & Punk #24 🤘🤘🤘 Newer »
This thread has been archived and is closed to new comments
| # | Наименование новости | Тональность | Информативность | Дата публикации |
|---|---|---|---|---|
| 1 | Quantum-inspired math could help AI recognize when it does not know the answer | 0 | 8.77 | 30-09-2026 |
| 2 | This robotic lawn mower makes sense in autumn too: Anthbot N8 review | 0 | 4.74 | 03-10-2026 |
| 3 | AI Confidence Starts with Packet Truth | 0 | 8.16 | 01-10-2026 |
| 4 | Власти США запустили свой чат-бот с функциями ИИ: он опровергает даже заявления Трампа | 0 | 19.59 | 04-10-2026 |
| 5 | The 'War Games' problem: Computer science has long understood what it takes to keep AI under control | 0 | 9.27 | 30-09-2026 |
| 6 | ИИ и блокчейн терпят неудачи без реальных задач пользователей | 0 | 10 | 12-09-2026 |
| 7 | heise-Angebot: Jetzt in der Make 5/26: Agentic Coding für Maker | 0 | 18.47 | 25-09-2026 |
| 8 | There’s One Word That Your Chatbot Should Never Say | 0 | 10 | 29-09-2026 |
| 9 | OpenAI отказалась выпускать ИИ-модель GPT-6.1 Astra. Есть проблемы | 0 | 12.11 | 30-09-2026 |