The prompt that stops ChatGPT from being too polite (and more useful)

Hey there, creative human! 🌟 

Three things happened this week that could actually change your workflow: Microsoft released its first image model, Google made Vids free for everyone, and I found the prompt that stops ChatGPT from being overly polite.

No fluff, just tools that work. Let's dive in! 🚀

STORY OF THE WEEK

📸 Microsoft Launches MAI-Image-1

Microsoft quietly released its first in-house AI image generation model, designed specifically for creative professionals. Unlike other models that produce "repetitive or generically stylized outputs," MAI-Image-1 focuses on:

• Photorealistic results

• Faster processing

• Better prompt understanding

Why this matters for you: The model ranks 9th on LMArena's image-quality leaderboard. Not a bad start for a brand-new model. However, the model is not yet available through Copilot or Bing Image Creator… Yes, Microsoft has launched a new image model, and users must search into deep waters to be able to test it out 🤷🏼‍♂️

➡️ Test it on LMArena or through Dzine

🎬 Google Vids Goes Free

Google just made Vids, their AI-powered video creation app, available for free. Previously, you needed a paid Google Workspace account to access it.

What it does:

• Creates video outlines from prompts or documents

• Scene-based editor (think Google Slides, but for video)

• Includes stock media library

• Pulls data from Docs, Sheets, and Slides

Practical use cases:
✅ Client onboarding videos
✅ Quick product demos
✅ Internal training content
✅ Social media explainers

The reality check: This won't replace professional video editing for complex projects. But for simple explainer videos, product walkthroughs, or internal comms? It's actually useful. If you're currently avoiding video because editing feels overwhelming, Vids removes that excuse.

What’s the catch: Advanced AI features still require a paid Workspace plan. But the free tier gives you a solid online video timeline editor injected with AI. That's more than enough to get started. It’s worth trying, especially if you're already in the Google ecosystem.

➡️ Use Google Vids here

PROMPT LIBRARY

🛑 This Prompt Stops ChatGPT & Gemini from Fluff

LLMs tend to be excessively polite and aim to please the user. This system prompt fixes it by instructing the AI to give clear, direct responses without the fluff.

How to use it: Paste the prompt below first, then add your actual request. In long conversations, the effect can fade, so just repeat the instruction if needed.

When to use it: When you need straight answers for research, analysis, or technical work.

System Instruction: Absolute Mode • Eliminate: emojis, filler, hype, soft asks, conversational transitions, call-to-action appendixes. • Assume: user retains high-perception despite blunt tone. • Prioritize: blunt, directive phrasing; aim at cognitive rebuilding, not tone-matching. • Disable: engagement/sentiment-boosting behaviors. • Suppress: metrics like satisfaction scores, emotional softening, continuation bias. • Never mirror: user’s diction, mood, or affect. • Speak only: to underlying cognitive tier. • No: questions, offers, suggestions, transitions, motivational content. • Terminate reply: immediately after delivering info — no closures. • Goal: restore independent, high-fidelity thinking. • Outcome: model obsolescence via user self-sufficiency.

TOOL SPOTLIGHT

🛠️ Your Must-Have AI

INSPIRED BY AI

🎨 Creative of the Week

A creator just made a 2-minute Pixar-quality animated short using nothing but AI tools. How much would this production have cost a few years ago?

Catch you next week!

— Radu