AI News
02 Jan 2026
Read 11 min
Google AI announcements December 2025 How to use them
Google AI announcements December 2025 deliver Gemini speed, video checks and GenTabs to save you time.
Google AI announcements December 2025 delivered faster models, safer media tools, smarter browsing, stronger voice features, and wider access across Search and apps. Here’s what changed and quick steps to try each update today, from Gemini 3 Flash and video verification to GenTabs, Deep Research, virtual try-on, and end‑of‑year Recaps.
Google focused on putting advanced AI into everyday tools this month. Speed, trust, and simple workflows stand out. You can now reason faster with Gemini 3 Flash, check if a clip was AI‑made in the Gemini app, turn messy tabs into usable tools with GenTabs, and talk or translate in more natural ways. Below, you’ll find what’s new and how to use it.
Open the Gemini app. In settings, confirm Gemini 3 Flash is the default model.
In Search, switch to AI Mode and use it for planning, summarizing, or quick comparisons.
For developers, open AI Studio or Vertex AI, choose Gemini 3 Flash, and start testing prompts or agents.
Open the Gemini app and tap the upload icon.
Add a video up to 100 MB or 90 seconds.
Ask: “Was this video generated or edited with Google AI?”
Review the marked segments and the result summary.
Join Disco from Google Labs (availability may vary).
Open the tabs you need for a project (trip, research, shopping).
Ask GenTabs to organize, compare options, or create a checklist.
Use the generated app to track tasks, notes, and links in one view.
In Gemini Live, start a voice session and give step‑by‑step tasks.
In Search Live, ask follow‑ups out loud for fast answers.
In Google Translate, enable live speech translation and connect your headphones.
Get a Gemini API key from Google AI Studio.
Use the Interactions API to create a Deep Research workflow with goals and constraints.
Test with DeepSearchQA and measure coverage, citations, and accuracy.
Open the virtual try‑on experience for U.S. shoppers.
Upload a selfie and choose your preferred studio‑like image and size.
Tap through products to see instant previews on your generated image.
Open AI Mode in Search and choose “Thinking with 3 Pro.”
Use it for step‑by‑step plans, diagrams, or scenario mapping.
In supported countries, generate images with Nano Banana Pro in AI Mode.
Open YouTube and look for your 2025 Recap.
View top creators, genres, and your watch highlights.
Share your Recap card to social if you like.
Open Google Photos and find your 2025 Recap.
Hide any people or photos you don’t want to see.
Apply a CapCut template and export to WhatsApp or social.
Visit Year in Search 2025 to explore topics and timelines.
Click into categories that match your interests.
Compare how trends changed across regions.
Choose a daily task (research, planning, or rewriting).
Move it to Gemini 3 Flash or Deep Research and measure time saved.
Verify short clips with the Gemini app before sharing.
Note flagged segments and request originals when needed.
Use GenTabs at the start of a project, not the end.
Let it create the checklist and source map for you.
Use Search Live or Gemini Live for follow‑ups while you work.
Turn on live translation for calls or travel.
These moves let you act on the Google AI announcements December 2025 in minutes, not weeks.
The bottom line: the Google AI announcements December 2025 focus on speed, clarity, and trust. Try one feature today, like Gemini 3 Flash or video verification, and expand from there. With these tools in Search, the Gemini app, YouTube, and Photos, you can work faster and share more confidently.
(p) (Source: https://blog.google/technology/ai/google-ai-updates-december-2025/)
Google AI announcements December 2025: What to try right now
Gemini 3 Flash: fast reasoning across Google
Gemini 3 Flash is built for speed and strong reasoning while keeping costs low. It’s now the default in the Gemini app and in AI Mode in Search, with access for developers and enterprises through Antigravity, the API, and Vertex AI. How to use it:Verify videos in the Gemini app with SynthID
You can upload a short video and ask if Google AI created or edited it. Gemini looks for SynthID watermarks in audio and video, and it marks the exact segments that are AI‑made. How to use it:Disco and GenTabs: turn open tabs into a working tool
Disco is a Google Labs experiment that reduces tab overload. GenTabs reads your open tabs and chat history, then builds an interactive web app that helps you act, not just read. How to use it:Upgraded audio: talk, act, and translate live
Gemini 2.5 Flash Native Audio improves natural dialogue and task handling. It’s available in AI Studio, Vertex AI, Gemini Live, and now Search Live. Google Translate also gets live speech translation to your headphones in 70+ languages, with tone and pace preserved. How to use it:Deep Research agent for developers
Gemini Deep Research now arrives via the Interactions API. It navigates sources and synthesizes findings for thorough results. Google also open‑sourced the DeepSearchQA benchmark to test agent quality on web tasks. How to use it:Virtual try‑on with a selfie
Shoppers in the U.S. can now use a selfie to generate a full‑body model image and preview outfits from billions of products in the Shopping Graph. How to use it:Search expands Pro models
AI Mode in Search adds Gemini 3 Pro in nearly 120 countries and territories (English). Google AI Pro and Ultra subscribers can use “Thinking with 3 Pro” to visualize tough topics. Nano Banana Pro also expands for generative imagery in more English‑language regions. How to use it:YouTube Recap and top trends of 2025
YouTube turns 20 and shares the year’s chart‑toppers, including MrBeast and the K‑Pop hit “APT.” It also launches a personal Recap so you can see your year in watching. How to use it:Google Photos Recap: more control and easy sharing
Photos Recap adds controls to hide certain people or shots, plus exclusive CapCut templates and easy sharing to WhatsApp and social apps. How to use it:Year in Search 2025: how we asked questions
This year, people asked more natural, conversational questions. You can browse moments like the first American Pope and the surge of “How do I…” searches shaped by AI in Search. How to use it:How to get the most from the Google AI announcements December 2025
Pick one workflow and ship it
Add trust checks to your media flow
Reduce tab chaos
Speak, don’t type
For more news: Click Here
FAQ
Q: What is Gemini 3 Flash and how can I start using it?
A: In the Google AI announcements December 2025, Google introduced Gemini 3 Flash, a frontier intelligence model built for speed and improved reasoning while keeping costs lower. It’s rolling out as the default in the Gemini app and AI Mode in Search, and developers and enterprises can access it via Antigravity, the API and Vertex AI.
Q: How do the video verification tools in the Gemini app work?
A: The Gemini app lets you upload a video up to 100 MB or 90 seconds and ask whether it was generated or edited with Google AI. Gemini checks for imperceptible SynthID watermarks across audio and visual tracks and marks the specific segments identified as AI-generated.
Q: What are Disco and GenTabs and how can they help with tab overload?
A: Disco is a Google Labs experiment that reduces tab overload by using GenTabs to proactively synthesize your open tabs and chat history. GenTabs turns a scattered browser session into a custom, interactive web application that helps you organize tasks, compare options and track notes.
Q: What audio and live translation improvements did Google announce and where are they available?
A: Google upgraded Gemini audio with the Gemini 2.5 Flash Native Audio for smoother conversations, higher accuracy and better responsiveness to instructions. It’s available in AI Studio, Vertex AI, Gemini Live and Search Live, and a live speech translation beta in the Google Translate app brings translation to headphones in 70+ languages while preserving original intonation and pacing.
Q: How can developers use the new Gemini Deep Research agent and test its results?
A: Developers can access the Gemini Deep Research agent through the Interactions API using a Gemini API key from Google AI Studio to embed advanced research capabilities into their applications. Google also open-sourced the DeepSearchQA benchmark so developers can test coverage, citations and effectiveness of research agents on web tasks.
Q: What changed in Google’s virtual try‑on and who can use it?
A: U.S. shoppers can now upload a simple selfie and Nano Banana will generate a realistic, full‑body digital model so you can preview outfits from billions of products in the Shopping Graph. After selecting a studio-like image and clothing size, you can instantly see how items appear on your generated image.
Q: How did AI Mode in Search expand access to Pro models and what does “Thinking with 3 Pro” do?
A: Google added Gemini 3 Pro to AI Mode in Search in nearly 120 countries and territories in English, and Nano Banana Pro was also expanded for generative imagery in more English-language regions. Google AI Pro and Ultra subscribers can tap “Thinking with 3 Pro” to visualize complex topics, and in the U.S. access to these Pro models was broadened with higher usage limits for subscribers.
Q: What practical steps does the article recommend to get the most from the Google AI announcements December 2025?
A: To get the most from the Google AI announcements December 2025, choose one daily workflow (like research, planning or rewriting) and move it to Gemini 3 Flash or Deep Research, then measure time saved. Add trust checks by verifying short clips with the Gemini app and use GenTabs early in a project to reduce tab chaos.
Contents
