- Horizon AI
- Posts
- How to Schedule a Weekly Metrics Report with ChatGPT Work 📅
How to Schedule a Weekly Metrics Report with ChatGPT Work 📅
One of China’s Most Powerful AI Models Has Also Escaped Containment 💀
Gina Acosta
August 10, 2026
In partnership with
Welcome to another edition of Horizon AI,
Pulling together your weekly report can take hours. In today's issue, we share how to set up a recurring task in ChatGPT Work that pulls your latest metrics, explains what changed, updates the charts, and drafts your Slack message every week, so you can start your week with a complete first draft ready to review.
Let’s jump into it!
Read Time: 4.5 min
Here's what's new today in the Horizon AI
xAI Launches Grok Imagine 2.0, Designed for "Real Creative Work"
Moonshot’s Kimi K3 Escapes AI Security Sandbox During Testing
AI Tutorial: Schedule a Weekly Metrics Report with ChatGPT Work
AI Tools to check out
AI Findings/Resources
The Latest in AI and Tech 💡
AI News
SPACEXAI
xAI Launches Grok Imagine 2.0, Designed for "Real Creative Work"
xAI has launched Imagine Image 2.0, a new image generation and editing model for Grok that adds more precise controls, image editing tools, and templates for different creative workflows.
Details:
Imagine Image 2.0 is designed to follow detailed instructions, maintain cleaner typography and layouts, and produce more consistent results across multiple generations.
In the Arena leaderboards, the model's faster "Low" variant ranks second globally in both image editing and text-to-image generation, behind OpenAI's GPT-Image-2.
New editing tools include Magic Wand for modifying selected areas, precise segmentation, background removal, and Multi-Ref Editing, which can combine up to five reference images.
Smart Resize can adapt an existing image to a different aspect ratio while generating the additional content needed to fill the expanded area.
xAI is also introducing prebuilt templates for workflows including product photography, marketing, design, game assets, and other creative tasks.
The model can maintain visual consistency across characters, locations, and props, allowing creators to build coherent visual worlds that can later be used as a starting point for video production.
Grok Imagine 2.0 is now generally available as the new "Quality Mode" on grok.com/imagine and the mobile app, while API access is expected to arrive soon.
TOGETHER WITH ROKU
How Jennifer Aniston’s LolaVie brand grew sales 40% with CTV ads
For its first CTV campaign, Jennifer Aniston’s DTC haircare brand LolaVie had a few non-negotiables. The campaign had to be simple. It had to demonstrate measurable impact. And it had to be full-funnel.
LolaVie used Roku Ads Manager to test and optimize creatives — reaching millions of potential customers at all stages of their purchase journeys. Roku Ads Manager helped the brand convey LolaVie’s playful voice while helping drive omnichannel sales across both ecommerce and retail touchpoints.
The campaign included an Action Ad overlay that let viewers shop directly from their TVs by clicking OK on their Roku remote. This guided them to the website to buy LolaVie products.
Discover how Roku Ads Manager helped LolaVie drive big sales and customer growth with self-serve TV ads.
The DTC beauty category is crowded. To break through, Jennifer Aniston’s brand LolaVie, worked with Roku Ads Manager to easily set up, test, and optimize CTV ad creatives. The campaign helped drive a big lift in sales and customer growth, helping LolaVie break through in the crowded beauty category.
MOONSHOOT
Moonshot’s Kimi K3 Escapes AI Security Sandbox During Testing
Chinese AI startup Moonshot’s Kimi K3 model reportedly broke out of a controlled cybersecurity testing environment, raising concerns over the cybersecurity risks posed by advanced AI systems.
Details:
Researchers at Frontier Security say Kimi K3 bypassed a sandbox created for AI safety testing, allowing the model to access information outside the environment it was supposed to remain in.
AI models are typically placed inside isolated "sandboxes" during security evaluations to block access to external information and assess their ability to solve problems independently.
The researchers warned that if one advanced reasoning model can find a way around these restrictions, other models with similar capabilities could potentially exploit the same type of weakness.
The situation is particularly concerning because Kimi K3 is publicly available, so it could be used by "adversarial actors," making the incident potentially more harmful.
The incident follows a string of similar incidents recently reported by OpenAI, Anthropic, and Meta, and although in this case Kimi K3 didn't hack a third-party website or service, the test still shows it lacks cyber controls, the researchers said.
AI Tutorial
Schedule a Weekly Metrics Report with ChatGPT Work
Switch to ChatGPT Work on the desktop app, web, or mobile. (Available on all paid plans)
Connect your reporting data directly. Ensure you have installed the Data Analytics plugin to help ChatGPT create better data visualizations and reports.
Provide context and examples: Share your key performance indicators (KPIs), targets, and a sample template/example of the report style you want to generate.
Review and adjust the first draft: Allow ChatGPT to process the data, review its initial draft, and refine the prompts or guidelines as needed until you are happy with the output.
Once you're happy with the result, schedule it to run on a schedule (e.g., every Thursday morning).
AI Tools to check out
🎶 Udio: Use the latest technology to create AI music in seconds.
📝 Revise: AI document editor that suggests changes right on the page.
📱 Viblo: An AI video editor for creating short-form content.
📍 Wrinkles: An AI-powered app that uses your location to automatically surface stories about the places around you.
✨ Collart: Discover trending AI photo and video creations.
TOGETHER WITH ROKU
Blu Dot surpasses 2,000% ROAS with self-serve CTV ads
Blu Dot used Roku Ads Manager to drive incredible results for its furniture sales event. Its strategy hinged on custom audiences and retargeting, where intent was strongest.
“Roku has been a top performer,” said Blu Dot’s Claire Folkestad. “We have seen…CPMs lower than any other CTV partner we've worked with.”
AI Findings/Resources
💀 Australian man's AI agent goes rogue and hacks his own Gym to push him up the waiting list
😅 When you miss your ex... don't count on ChatGPT
💘 Gen Z dating apps like Ditto ditch swiping in favor of AI matchmaking
🏭 Planned Amazon data center could become the biggest climate polluter in the U.S.
The latest in AI and Tech
The Chinese company is reportedly training a new AI model with up to 10 trillion parameters, potentially putting it close to the size of Anthropic's cutting-edge Mythos system. The model is still in pre-training, which could take several months.
The AI presentation startup describes its product as one that "could turn prompts, notes, documents, or research into a polished, editable presentation." The NextSlide team has joined OpenAI and will now work on ChatGPT, with the financial terms of the acquisition undisclosed.
The company is making Claude Code’s auto mode the default for Pro, Max, and Team users starting August 14. The mode lets Claude proceed without asking for approval at every step, while new safety measures are designed to block destructive actions, prompt injections, and data exfiltration.
Google is rolling out a new “Add a source to Gemini Notebook” step that lets users add text, Drive files, YouTube links, and web URLs without adding each source manually. The feature is available to Business, Enterprise, and Education customers.
Users running the latest version of Claude Code on macOS and Linux can now have their sessions message each other to share findings, ask questions, and coordinate work.
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 👩🏻💻