Google I/O kicks off today, and if the company’s recent moves are any indication, we’re about to hear a lot about AI woven into the fabric of your daily life. From health tracking to laptop operating systems, Google is making a calculated bet that Gemini can become the invisible assistant that actually gets things done.
But before we get excited about AI agents that book appointments and manage your medical records, there’s something more pressing that deserves attention: whether anyone actually trusts these companies with their most sensitive data.
The Health Data Gamble
Google’s latest health announcements paint a clear picture of where the industry is heading. The rebranded Fitbit app, now called Google Health, pairs the company’s new displayless Fitbit Air tracker with an AI health coach powered by Gemini. This isn’t just about counting steps anymore. The system can sift through years of biometric data and medical records, then offer personalized training advice based on your complete health history.
It’s genuinely useful technology. Companies like Whoop and Oura are already building similar AI health coaches, so Google isn’t alone in recognizing the potential. But here’s the catch: health data has a long and messy history of being exposed, shared, or sold without permission. Even companies with strong privacy promises have failed to protect it.
Letting Gemini draft an email is one thing. Letting it interpret your medical history is something entirely different.
The Assistant Evolution
Beyond health, Google is doubling down on making Gemini a genuinely capable assistant across your devices. At last week’s Android Show, the company previewed Android 17 and what it’s calling “Gemini Intelligence.” This new iteration of the assistant goes beyond voice commands. It can handle smart scheduling, autofill documents using data pulled from across Google apps, and generate custom widgets that refresh automatically.
The vision is compelling: AI agents that bridge the gap between simple requests and complex multi-app tasks. But Google has been vague about how this actually works in practice. The company showed a handful of examples but didn’t clearly explain how Gemini Intelligence will handle the messy reality of jumping between different apps and services while remembering context.
This is where Technology roadmaps often diverge from real-world usability. The promise of AI agents has been circulating for years. Today at I/O, we’ll finally get a sense of whether Google has actually cracked the problem or is still a few iterations away.
Googlebooks and the Chromebook Replacement
Google also unveiled Googlebooks last week, positioning them as the next generation of laptops built on Android instead of Chrome OS. These aren’t actual products yet, but they signal a significant shift in how Google thinks about its desktop strategy. Gemini’s AI capabilities will be deeply integrated into the experience.
I/O might shed more light on what this actually means for developers and users. Is this just a rebranding effort, or does it represent a fundamental rethinking of how AI should work on your computer?
The Privacy Problem That Won’t Go Away
Here’s the uncomfortable truth: none of this works if people don’t trust these companies. Google can promise strong privacy protections, but the market has been flooded with those same promises before. When you’re asking users to hand over their medical records to an AI system, skepticism isn’t paranoia. It’s warranted caution.
This is the real test for Google at I/O. Not whether the technology is impressive, but whether the company can meaningfully address the trust deficit that exists around sensitive data.
The announcements will likely be technically sophisticated and strategically sound. But the harder question remains: will anyone actually believe they’re safe?
