- Horizon AI
- Posts
- You Can Now Turn Off the Visible Watermark on Your Gemini AI Creations. Here's How 👇
You Can Now Turn Off the Visible Watermark on Your Gemini AI Creations. Here's How 👇
OpenAI introduces "Computer History" to ChatGPT for Mac 🖥️
Gina Acosta
August 17, 2026
In partnership with
Welcome to another edition of Horizon AI,
In today's issue, we share how to remove the "sparkle" watermark that appears in the bottom-right corner of content generated with Google's Nano Banana and Omni models, making your output cleaner and more usable for professional and creative work.
Let’s jump into it!
Read Time: 4.5 min
Here's what's new today in the Horizon AI
ChatGPT’s New Computer History Tracks How You Use Your Mac
Google Launches Gemini 3.7 Flash for Coding and AI Agents
AI Tutorial: Google Gemini now lets you remove visible AI watermarks
AI Tools to check out
AI Findings/Resources
The Latest in AI and Tech 💡
AI News
OPENAI
ChatGPT’s New Computer History Tracks How You Use Your Mac
OpenAI has introduced Computer History for its ChatGPT desktop app on macOS, allowing the AI to remember your activity across apps and websites so it can better understand your workflows, suggest automations, and pick up unfinished tasks.
Details:
Computer History turns your activity across apps and websites into memories and a timeline that ChatGPT and Codex can reference when responding to requests.
The feature is opt-in, and users can exclude specific apps or websites from being tracked. They can also delete individual history entries for additional control.
OpenAI says the feature automatically ignores activity in private or incognito browser tabs.
OpenAI shared a demo video with Dominik Kundel, from its Developer Experiences team, showing the app looking up his last edited document, checking if it was shared via Slack, and recapping his morning.
The new feature has drawn comparisons to Microsoft's controversial Windows Recall feature, which relied on continuous screenshots of everything you do on your PC. But OpenAI says Computer History remembers what users have been doing across apps and websites using interaction events, such as clicks, typing, and app switching, rather than by capturing your screen or audio.
TOGETHER WITH PORKBUN
Domain Names + Web and Email Hosting Done Right
Tired of overpaying GoDaddy or Namecheap? There's a smarter option with Porkbun!
Porkbun is the domain registrar trusted by creators, developers, entrepreneurs, and anyone who wants low prices without the nonsense or the forced upsells.
Here's why so many people are making the move to the Bun:
• Most domains sold at cost
• Low, transparent registration and renewal pricing
• Free features like WHOIS privacy and SSL certificates
• Powerful web and email hosting options
• Real human support 24/7, 365 days a year
• Named the #1 domain registrar by Forbes Advisor and USA Today
Whether you're launching a business, building a personal brand, starting a side project, or creating your very first website, Porkbun makes the whole thing easy.
Google Launches Gemini 3.7 Flash for Coding and AI Agents
Google has introduced Gemini 3.7 Flash, a new Flash-series model focused on coding, software development, and agentic workflows.
Details:
Gemini 3.7 Flash improves coding performance, particularly for debugging, resolving issues, and generating production-ready software, with Google reporting higher scores on several software engineering benchmarks.
The model also performs better in web development, generating more complete applications with fewer prompts and improving its ability to follow reference designs, screenshots, and design systems.
For knowledge-intensive work, Google says 3.7 Flash shows stronger reasoning in areas such as finance, law, and biosciences, while also performing better on real-world business automation tasks.
The model is designed to be more agentic, with better multi-step planning, tool use, and instruction following. Google says this can reduce the need for manual oversight and repeated attempts when agents encounter problems.
Gemini Spark, Google's 24/7 personal AI agent, is also being upgraded to 3.7 Flash, improving its ability to work with Google Workspace apps and handle tasks such as organizing files, drafting emails, and updating documents.
To drive adoption, Google is offering Gemini 3.7 Flash at an introductory rate of 75 cents per million input tokens and $3.75 per million output tokens through the end of the year, half the original cost of Gemini 3.6 Flash.
AI Tutorial
Google Gemini now lets you remove visible AI watermarks
Users can now toggle watermarks off for images made with the Nano Banana AI model, videos made with Omni, and songs made with Lyria.
Go to Gemini and sign in
Click on settings
Toggle "Media Watermark" off to remove watermarks from all future images, videos, and music tracks.
The change only affects the visible marker. Google says AI-generated content will still include invisible SynthID watermarks and C2PA metadata, which can be used to identify AI-generated content.
AI Tools to check out
🎶 MiniMax Music 3.0: Next-generation open-weights, production-ready & versatile music model.
🔴 Loom: Easily record and share AI-powered video messages with your teammates and customers to supercharge productivity.
🔍 Huncher: Find the leaks in your site. Fix them with AI.
🚀 Acebuilder: Ship a production-ready landing page in minutes.
🧠 Mixpanel: Product Intelligence built for the AI era.
AI Findings/Resources
📌 One prompt tweak that's worth trying if you use Claude a lot
🌎 How small a window life on Earth has occupied, visualized with GPT-5.6
🌱 Nvidia-powered lasers enable herbicide-free weed control, paving the way for chemical-free agriculture
The latest in AI and Tech
Anthropic says Claude's new text watermark is "a version" of the open-source SynthID-Text system Google developed.
According to the company, light editing may not remove the watermark, while full rewrites will, since "the watermark only applies to words Claude chooses." For text that was only proofread or edited by Claude but written by a person, the presence of a watermark will depend on "the length of the text and how heavily Claude has edited it."
The company says its Qwen family of open-weight AI models has surpassed 3 billion downloads globally over the past six months. The company has released more than 460 models, with the ecosystem also generating over 300,000 derivatives as developers increasingly use Qwen as a base for new AI systems.
In its announcement, Cursor said that it will now have access to "the largest fleet of GPUs in the world," which will give it the computing capacity to build and train better models that it can offer its customers at a lower cost.
Anthropic CEO Dario Amodei pushed back on claims that he's been overly negative in his portrayal of AI and its potential impact on the future, thus fueling public opposition to the technology.
He argues that the deeper issue is a broader crisis of trust in companies, governments, and the tech industry. He also said AI companies need to deliver real benefits rather than relying on promises, while defending Anthropic's support for targeted regulation.
Axios and OpenAI have signed a three-year partnership to support AI-assisted local journalism, with OpenAI providing tools to 13 local newspapers and Axios making its news content available for training ChatGPT.
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 👩🏻💻