Hacker News Reader

Comment by infoyog | Hacker News Reader

10 points·infoyog··8 comments

Are these really hacks? This is just using ChatGPT normally and as intended.

I think the author's intended use of the word 'hack' is less about veering off the beaten path to get it to do something unexpected, but more inline with 'life hack' or 'cool things to try'. The intended audience for a post like that might be the less-technical crowd.

I was just going to write the same - not a hack, unless the definition of 'hack' has changed recently

It's interesting that many ChatGPT "prompt hacks" also work on other language models. I tried copying the DAN (Do Anything Now) prompt into the LLaMa 2 system prompt, and it worked more or less unchanged: https://twitter.com/thomasahle/status/1683508252821291008

Is this a real article?

> 9. Personal Assistant and Productivity Tool ChatGPT functions as a virtual personal assistant, aiding users in managing daily tasks and increasing productivity.

> Example:

> Prompt: "Set a reminder to buy groceries tomorrow at 5 PM."

This is obviously nonsensical. ChatGPT can't remind you of anything because it only responds when you ask it, and it doesn't know the time.

The only interesting piece on here is “Set a reminder.” How is that supposed to work?

Edit: I’m going to flag this and I suggest others do as well—it’s blogspam with a clickbait headline, and incorrect information in it. I’m pretty sure it’s written in GPT.

As someone who actually writes haiku, I hate the "write a haiku" ChatGPT example as it produces these stilted, lifeless eighth grade poems that people nevertheless copy-paste into their work.

Prompt: "Set a reminder to buy groceries tomorrow at 5 PM."

Response: "Reminder set: Buy groceries tomorrow at 5 PM."

Uh... what? All this does is yield the usual "As an AI language model, I can't set reminders on your device. But here's how to build an alarm clock..." response.

Reply on news.ycombinator.com