Day 1 of Google I/O’s 2025 edition has come to a close, and, as expected, it was jam-packed with paradigm-shifting announcements. Unsurprisingly, AI was a hot topic (though Sundar mentioned the term less than the ~120 times he dropped the buzzword in 2024).
While there were dozens of new products and features announced, we wanted to take a moment to highlight the standouts and key themes we heard and what we’re excited about coming out of the show.
Firstly, AI crept its way into all things, both obvious and more subtle. Major updates to Google Beam (formerly Starline) have AI baked in, helping bring the “in-person” feel to virtual meetings without any headwear. While not blatant, AI is the underlying technology that enables all of this to work. These under-the-hood AI capabilities make a strong case for AI proliferation into all applications, even those that are not primarily “AI-focused.”
That said, some of the announcements that got us the most excited were indeed explicitly AI-based.
Democratizing development: from idea to application with AI
What stood out most was how AI is democratizing the application development process. Whereas early incarnations of AI tools for developers simply aided them with things like code completion, this new generation of tooling represents a step-function increase in capabilities.
Take, for example, a talented architect who knows nothing about frontend development, design, or user experience. She may have great application ideas and understanding of the underlying architecture, but is completely dependent on other teammates to hone her ideas.
Leveraging only tools from today’s announcements, she can now move from idea to running application using only her imagination and AI assistants.
The AI-powered journey: a new workflow for creators
First, she can use Google Stitch to create a UI with emphasis on a great user experience. She can iterate, provide feedback, and get guidance from her AI agents built specifically to help developers think through this process.
With this complete, she can now export her designs to Figma, enabling sharing with her colleagues for feedback, tweaks, and enhancements. She’ll gain more than just static images or mockups – instead, she and her colleagues will be able to interact with the design and iterate in real time.
Once the work in Figma is complete, she can take her Figma files and provide them to Firebase Studio. Firebase Studio will then take those Figma files and create a functional prototype. Using only the provided files, the Gemini agent embedded in Firebase Studio will write all of the code to take the files and turn them into a functional application frontend.
Using Gemini Code Assist, she can now guide Gemini toward writing the functional backend services she has envisioned. Gemini will self-identify and self-correct errors and iterate through the development process in to generate a functional application. This takes the Figma UI and turns it into a functional, full-stack application. That’s not a typo. With the new announcements around Firebase Studio, the Gemini agent can now account for the backend as well, managing users, their identities, and application storage within Firebase.
Tasks that are often overlooked, like adding comments, writing clear readme files, or coding tasks that you just don’t want to do yourself, can be handed off to Jules, the asynchronous coding agent. Tell it what you want it to do and go back to your own tasks. Jules will start working on a plan via its reasoning engine and will notify you once the plan is ready for your review. Once you approve the plan, it will work on the tasks by itself.
To complete the process, the deep integration between Google Cloud Run and Firebase Studio now means that she can deploy her application to Google Cloud Run with a single click, deploying the app into production.
AI’s superpower: amplifying human capabilities
All of this can be done without code and can be as collaborative or as siloed an effort as she desires. And she gains the ability to collaborate at every step, whether with human colleagues or her generative AI agents.
This democratization of the development process means that more people from more backgrounds and with a greater variety of skills can now create full-featured applications with only their imagination as the limit.
This expansion of human capabilities was the real theme we picked up from I/O. It wasn’t that AI was positioned to take away human jobs or functions, but rather the AI enabled humans, turning them into superheroes. By leveraging these tools, more people are now able to accomplish more. This breaks down barriers, expands what individuals can do, and supercharges the power of dynamic teams.
What will you build?
With all of the announcements made today, how do you envision yourself making use of these tools to supercharge your capabilities?
Whether you’re looking to implement these new capabilities, navigate specific challenges, or strategize your next steps with AI, our team of experts is here to help. Don’t hesitate to reach out to us for a personalized discussion on how we can support your journey.
For executive leaders looking to gain a deeper, strategic understanding of the overarching forces shaping the AI landscape for their entire organization, download Google’s AI Business Trends 2025 Report. This essential report uncovers 5 critical AI shifts set to redefine business, providing actionable insights and expert perspectives to help you build a resilient and innovative AI strategy.
