STOP using custom instructions in ChatGPT

Futurepedia · Intermediate ·🧠 Large Language Models ·4mo ago

Key Takeaways

This video teaches viewers to use custom instructions in ChatGPT Projects instead of applying them to every chat

Original Description

Everyone is using custom instructions wrong in ChatGPT. You don’t want them to apply to every chat. Really, you want to be using them in Projects. You can keep every topic siloed with its own unique instructions so you context doesn’t bleed over into your other chats. #chatgpt #chatgpttricks

Watch on YouTube ↗ (saves to browser)

Sign in to unlock AI tutor explanation · ⚡30

Related Reads

LLMs Cannot Be Audited. In Compliance, That Is the Problem.

LLMs are being used for tax and compliance queries, but their lack of auditability poses a significant problem

Medium · Data Science

Don't Fragment My AI Stack: Why Shutting Off Chinese Open-Weight Models Is a Bad Idea

Learn why shutting off Chinese open-weight models can fragment AI stacks and hinder innovation, and how developers can advocate for open access to AI models.

Dev.to · Hanzla Baig

Marking the Wrong Symptoms: Evaluating LLM Watermarks in Medical Texts

Learn how LLM watermarks impact medical texts and why evaluating them is crucial for reliable traceability in clinical workflows

ClickGuard: Detecting and Spoiling Clickbait News with Informativeness Measures and Large Language Models

Learn how ClickGuard uses large language models and informativeness measures to detect clickbait news and improve online browsing experience

5 Levels of AI Agents - From Simple LLM Calls to Multi-Agent Systems

Dave Ebbelaar (LLM Eng)