- Horizon AI
- Posts
- Google’s New AI Model Can Use a Web Browser Like You Do 🖱️
Google’s New AI Model Can Use a Web Browser Like You Do 🖱️
Create AI Videos for Free with Grok Imagine
Gina Acosta
October 08, 2025
In partnership with
Welcome to another edition of Horizon AI,
Google has just introduced a new AI model that can surf the web for you, click buttons, and fill out forms.
Let’s check it out!
Read Time: 4.5 min
Here's what's new today in the Horizon AI
Google’s Gemini 2.5 Computer Use Model Can Navigate the Web Like a Human
European Commission Launches New AI Strategies
AI Tutorial: Create AI Videos for Free with Grok Imagine
AI Tools to check out
AI Findings/Resources
The Latest in AI and Tech 💡
AI News
Google’s Gemini 2.5 Computer Use Model Can Navigate the Web Like a Human
Google is previewing a new Gemini 2.5 Computer Use model that leverages 'visual understanding and reasoning capabilities' to surf the web on your behalf, retrieving information, filling out forms, and taking actions on websites, all from a single user prompt.
Details:
It operates in a continuous loop: the system captures a screenshot of the environment, the user’s request, and a record of previous actions, then generates the next UI action. After each action, a new screenshot is sent back to the model, and the cycle repeats.
It currently supports 13 actions, including opening a web browser, typing text, and dragging and dropping elements. Unlike alternatives from OpenAI and Anthropic, however, Google’s model only has access to a browser, not an entire computer environment.
Google claims the model “outperforms leading alternatives on multiple web and mobile benchmarks,” but also acknowledges that “it is not yet optimized for desktop OS-level control.”
Gemini 2.5 Computer Use is available through Google AI Studio, Vertex AI, and as a demo on Browserbase.
Google is stepping into the race with other leading AI labs that seek to extend their models into 'agents' capable of taking more actions on behalf of users across websites. It will now have to compete with OpenAI's ChatGPT Agent and Anthropic's Computer Use, both released over the past two years.
TOGETHER WITH TYPELESS
Typing is a thing of the past
Typeless turns your raw, unfiltered voice into beautifully polished writing - in real time.
It works like magic, feels like cheating, and allows your thoughts to flow more freely than ever before.
With Typeless, you become more creative. More inspired. And more in-tune with your own ideas.
Your voice is your strength. Typeless turns it into a superpower.
EUROPEAN UNION
European Commission Launches New AI Strategies
The European Commission rolled out two linked plans, Apply AI and AI in Science, to push AI into industry, the public sector and research.
Details:
Backed by about €1B now and bigger boosts under Horizon Europe, the goal is to make Europe “AI first” in practice, not just policy.
Apply AI targets real deployments in sectors like healthcare, manufacturing, defense, media, and mobility, with pilots, shared infrastructure, and direct support for SMEs through upgraded European Digital Innovation Hubs turned into Experience Centres for AI that offer hands-on help and connections to EU networks.
AI in Science centers on RAISE, a virtual institute that pools resources and expertise to advance AI-driven science.
This marks a major step toward Europe’s goal of establishing itself as a global hub for safe, innovative, and industry-driven AI, reinforcing its position in the global AI race.
AI Tutorial
Create AI Videos for Free with Grok Imagine
xAI recently released version 0.9 of their Imagine video generation model, featuring major upgrades in visual quality, motion, audio generation, and more.
And it’s free for all users, at least for now.
Go to Grok.com or open the mobile app.
Select Imagine.
Upload any photos.
Choose between Normal, Fun, or Custom mode if you want to add a specific prompt.
After a few minutes, your video with sound will be generated. You can select Redo to try a different mode (Normal, Fun, or Custom).
Download or share your video.
The possibilities are endless, one popular use is bringing old photos to life.
AI Tools to check out
🍋 LangLime: A language learning app that helps you learn 4× faster through real-world immersion.
🦾 Instruct: Automate anything with a single prompt.
👉 Everyday: Get tasks done across apps with plain English.
🍊 Fruitful: Track competitor websites and capture every change.
📹 Mosaic: A canvas for agentic video editing.
AI Findings/Resources
📚 Learn to build effective Agentic AI systems with Andrew Ng
🦾 Figure humanoid robots have now been running on the BMW X3 body shop production line for five months
👀 12-Factor Agents: Patterns of reliable LLM applications — Dex Horthy, HumanLayer
🧊 A user connected Claude with Blender, and now it’s generating wild 3D scenes
The latest in AI and Tech
The company is rolling out its AI-powered search mode to over 40 new countries and 35 languages, covering much of Europe.
The expansion will complete within a week, bringing the feature to more than 200 countries worldwide.
Anthropic is partnering with IBM to integrate its Claude models into IBM’s developer tools and software, starting with its latest IDE for enterprise engineers.
The plan is expanding to 36 more countries. Check the full list here.
It offers benefits such as higher limits for our image generation and editing model, Nano Banana, as well as increased access to video generation in the Gemini app, Flow, and Whisk.
It’s the company’s latest and most advanced vision-language model, offering multilingual and multimodal understanding and reasoning capabilities. It currently ranks No. 3 in LM Arena’s Vision Arena.
Google’s AI vibe-coding app is now available in more countries, including Canada, India, Japan, Brazil, and Singapore. There are also new improvements to Opal’s core performance and its debugging program.
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 👩🏻💻