Google’s Gemini AI is rolling out a wave of upgrades, making its research capabilities sharper, its responses more personal, and its connected apps more powerful — all at no cost to try. Whether you’re tackling a deep research project, streamlining your daily tasks, or just looking for an AI assistant that actually remembers your preferences, Gemini’s latest features aim to make AI feel less like a chatbot and more like a real assistant.
Smarter Research, Less Effort
First up, Deep Research, Gemini’s AI-powered research assistant, is now available to everyone in over 45 languages. Originally introduced in December, it was designed to save hours by gathering, analysing, and synthesising information from across the web. Now, it’s been supercharged with Gemini 2.0 Flash Thinking Experimental, making it even better at reasoning, analysing, and building multi-page reports. Plus, for the first time, you can see Gemini’s thought process as it works through research queries — a small but significant step toward making AI less of a black box.
If you’re a Gemini Advanced user, you’ll also get access to a 1 million token context window on 2.0 Flash Thinking Experimental. In simple terms? The AI can now handle and analyse much larger chunks of information at once, making it possible to tackle far more complex projects.
AI that actually feels personal
The AI assistant experience has always been a bit impersonal — until now. Gemini is introducing Personalisation, an experimental feature that taps into your Google apps (starting with Search) to provide responses tailored to your preferences. Ask for restaurant recommendations, and it will reference your recent searches. Need travel advice? Gemini will factor in the destinations you’ve been looking at. And if you’re worried about privacy, you remain in full control—you can disconnect it from your search history whenever you want.
More seamless app integration
If you’ve ever wished your AI assistant could handle more complex requests without needing multiple prompts, Gemini’s latest update has you covered. With the new Connected Apps feature, Gemini now integrates with Calendar, Notes, Tasks, and Photos. That means you can ask it to find a cookie recipe on YouTube, add the ingredients to your shopping list, and find an open grocery store — all in one go.
And soon, Google Photos will be part of the mix, allowing Gemini to pull insights from your pictures. Imagine asking it to build an itinerary based on past trips or reminding you when your driver’s license expires based on a photo.
Build your awn AI with Gems
For those who want even more control, Gems is rolling out to all users, allowing you to create custom AI personas tailored to specific needs. Want a personal nutritionist, a coding tutor, or a study buddy? With Gems, you can build a specialised AI with just a few instructions, and even upload files for added context.
The bottom line
Gemini’s latest updates aren’t just incremental improvements — they’re a much needed step toward making AI more useful, intuitive, and actually personal. Check them out at gemini.google.com.