- Horizon AI
- Posts
- Add This One Sentence to Your ChatGPT Prompts for Better Results 📝
Add This One Sentence to Your ChatGPT Prompts for Better Results 📝
Meta Launches Muse Image, Its New AI Image-Generation Model 🎨
Gina Acosta
July 08, 2026
In partnership with
Welcome to another edition of Horizon AI,
One of the biggest reasons AI gives disappointing answers is that it's often forced to guess what you mean. Instead of asking for clarification, ChatGPT will often just make assumptions.
In today's issue, we share a simple prompt you can add that encourages the AI to ask questions first, so you get more accurate and useful responses.
Let’s jump into it!
Read Time: 4.5 min
Here's what's new today in the Horizon AI
Meta Launches Image Generation Model with Coding, Search Capabilities
China Reportedly Considers Export Restrictions on Advanced AI Models
AI Tutorial: This One Sentence Can Instantly Improve Your ChatGPT Responses
AI Tools to check out
AI Findings/Resources
The Latest in AI and Tech 💡
AI News
META
Meta Launches Image Generation Model with Coding, Search Capabilities
Meta has introduced Muse Image, the first image generation model built by the company's Superintelligence Labs.
Details:
Muse Image uses an AI agent approach, calling tools to produce more accurate results and refine its own outputs.
Those tools include writing and running code to generate diagrams, scannable QR codes, animated GIFs, websites, or interactive games, as well as a web search feature that grounds images in current facts and real-world references, improving accuracy on knowledge-heavy prompts
On Image Arena, it ranks second in human preference scores for text-to-image generation and for both single- and multi-image editing, trailing OpenAI's GPT Image 2 in each category but beating models like Nano Banana and Grok Imagine.
One of its new features allows users to reference public Instagram accounts in prompts, enabling the model to generate new images using publicly available photos from those profiles. Users can opt out by changing their Instagram settings.
Muse Image is available through the Meta AI app, Instagram Stories, WhatsApp, and meta.ai, with additional integrations planned for other Meta products. The company also announced that a new 'Muse Video' model is in development.
TOGETHER WITH LEVANTA
7 Proven External Traffic Strategies
Most eCommerce brands running external traffic aren't scaling — they're just spending.
Wrong channels, no real attribution, and at the end of the month, still no clear answer to the question that matters: what actually drove revenue?
The brands getting it right aren't necessarily spending more. They've just stopped guessing. They know which channels pull weight on Amazon listings, which ones bleed budget, and why affiliate and creator traffic outperforms on ROI when it's set up correctly.
Levanta put together a free playbook breaking down 7 proven external traffic strategies — where each one works, where it falls apart, and what it takes to scale without it becoming a second job.
Inside you'll see how top brands are driving millions in off-Amazon revenue and why most channels underdeliver when brands don't know what to look for.
If you're serious about growing outside of PPC, this is worth 5 minutes.
AI RACE
China Reportedly Considers Export Restrictions on Advanced AI Models
China is reportedly considering new restrictions on access to its most advanced AI models, a move that could reshape global AI competition and make it harder for companies outside the country to use leading Chinese models.
Details:
According to Reuters, Chinese officials have held discussions with major AI companies, including Alibaba, ByteDance, and Z.ai, about limiting foreign access to advanced and future AI models.
The proposals reportedly include tighter controls on frontier AI systems, possible security reviews for advanced models, and new rules governing overseas investment in Chinese AI startups.
The discussions come as Chinese models such as Qwen, Doubao, and GLM-5.2 have gained international attention for offering strong performance at lower costs.
The reported plans mirror recent U.S. restrictions on Anthropic's latest advanced AI models, highlighting how both countries are increasingly treating frontier AI as a strategic national asset.
AI Tutorial
This One Sentence Can Instantly Improve Your ChatGPT Responses
Go to ChatGPT and write your normal prompts as you usually would.
Add this sentence to the end:
Before answering, ask up to three clarifying questions only if they're necessary to avoid making incorrect assumptions. Otherwise, answer immediately and mention any assumptions you made.If the AI needs more information, answer its clarifying questions before it generates the final response. If it already has enough context, it'll answer immediately while explaining any assumptions it made.
You can also try this in Claude, Gemini, or another AI chatbot.
This prompt works best for tasks where there are multiple possible answers, such as writing emails, planning trips, brainstorming ideas, creating workout plans, coding projects, or making important decisions.
AI Tools to check out
🔥 ZCode: The official development environment for GLM-5.2. Combines the best AI agents with your existing tools so you can plan, code, review, and deploy without friction.
🔮 Futuresearch: Forecasting backed with a public track record on stocks, prediction markets, public benchmarks, and forecasting tournaments.
🏴☠️ The Hugging bay: The Pirate Bay for open LLMs — model weights downloadable via torrents.
🦾 Jira: A powerful project management tool that helps teams plan, organize, and automate their work.
💼 Badge: AI agents collect peer reviews to generate proof of work.
AI Findings/Resources
😅 Someone made a song called “Claude’s Plan” inspired by Drake’s God’s Plan
👀 Anthropic shares a strategy to offset Fable 5's high cost, suggesting its use as an 'advisor' that provides guidance to a faster, lower-cost executor model, which handles the majority of the work
🔨 Build your own AI agent from scratch using the Vercel AI SDK, Composio, Supermemory, Telegram, and Vercel Cron
🚨 Chinese AI models are gaining ground with U.S. companies as OpenAI, Anthropic costs surge
The latest in AI and Tech
OpenAI has received approval from the Trump administration to publicly launch its GPT-5.6 model family on Thursday. The lineup includes Sol, the flagship model, Terra for high-volume workloads, and Luna, a faster, lower-cost model for everyday tasks.
Meta is reportedly testing AI-powered smart glasses with an always-on 'Super Sensing' mode that continuously captures audio and images throughout the day, allowing users to ask AI about things they've seen or heard. The prototype has sparked internal privacy concerns, and although Meta has not confirmed the project, it said it remains committed to privacy-focused technology.
Chinese AI startup MiniMax is reportedly preparing to release a 2.7 trillion parameter large language model as open source later this year. The model, internally known as M3 Pro, would be the largest Chinese AI model to date and could launch as early as the third quarter.
OpenAI's chief futurist, Joshua Achiam, has announced he is leaving the company after nine years. In a post on X, Achiam said there was no single reason for his departure, adding that he believes he can continue contributing to the safe development of AGI from outside a frontier AI lab.
Google DeepMind has introduced four new capabilities for Managed Agents in the Gemini API, including background execution, support for remote Model Context Protocol (MCP) servers, custom function integration, and credential refresh without resetting the agent's state.
That’s a wrap!
We'd love to hear your thoughts on today's email!Your feedback helps us improve our content |
Not subscribed yet? Sign up here and send it to a colleague or friend!
See you in our next edition!
Gina 👩🏻💻