Deep Research 🤝 Gemini Live. Use your voice to explore a topic — in depth — with Gemini Live's Deep Research integration. 1. Just ask the Google Gemini to run a Deep Research report on a topic, then feel free to close the chat, lock your screen, or keep... chatting about other things. 2. Gemini will work asynchronously in the background and send you a notification when your full research report is ready. 3. Once it is, you can chat through the results, ask follow-up questions, or refine the details.show more

Learn more:

@GeminiApp Hey @Google, will you celebrate Dino’s birthday on September 23rd? @googlechrome @ChromeDinoRH

@GeminiApp Please add project folders to Gemeni Spark 🙏. Gemeni Spark is really good but organize the chats would be very helpful 👏👏👏

@GeminiApp Y para cuando la búsqueda web en la app? Para cuándo vamos a poder elegir que la IA haga una consulta rápida en internet para evitar que alucine y responda cosas inventadas?

@GeminiApp I wonder where"Israel"is even mentioned in the Kalshi and Polymarket Intel tweet that caused Google to translate it as "Israel"?I’ve noticed strange changes to sentences and news reports several times.And they all point in the same direction.Mostly in the search engine's AI mode.

@GeminiApp background mode is clutch

@GeminiApp ❌

@GeminiApp So I can lock my phone, touch grass, and let an AI do the overthinking for me. Dangerous for my vintage archive habit, honestly.

@GeminiApp En algunos casos , Gemini Lite y sus versiones integradas en app se sale de contexto en el mismo chat , hay que mejorar más eso

@GeminiApp Cool idea, but I hope the info isn’t just a quick Google rehash. I’d want sources and real depth if I’m trusting it for research. Anyone tried it for something super obscure yet?

@GeminiApp HOW DO I RECOVER A GMAIL PASSWORD IF MY PHONE ISN’T WORKING. THERE’S NO OTHER OPTION TO VERIFY

@GeminiApp How are citations handled in voice mode? Does it read sources out, or do you have to open the written report for them?

@GeminiApp For the voice follow-up, “which source supports that?” needs to be as easy as “tell me more.” An audio summary is convenient, but the path back to the cited passage still matters when a finding is surprising.

@GeminiApp GEMIMI语音对话模型有几下严重不足 缺陷 1:语音对话模型缺乏思考推理逻辑链深层语意理解推理 2:语音对话模型视频缺乏深层推理思考 3:语音对话模型目前是个词语表面化答案简短空洞应答机器 期望优化改善语音对话模型笨没有思考推理没想法没逻辑问题

@Google @Gmail @googleaccount Being able to start a research task by voice, let it run in the background, and then discuss the findings sounds genuinely useful. As someone who has worked across several industries, I can see myself using this to explore unfamiliar topics and ask follow-up questions without interrupting my work. I’d like to try it, but I’m still unable to access my Google account. My only Google account was disabled for alleged multiple-account abuse or possible automated creation. I believe this is a mistake. I explained my situation in my appeals, but the rejection messages did not address my explanation. I have now reached the appeal limit. During recovery attempts, I also repeatedly encountered “Too many failed attempts” at the phone/SMS verification step, even after waiting several days. Several paid services linked to this account are now difficult or impossible for me to access and manage. Could you please help me reach a human account specialist who can review whether my case can be reopened? I can provide screenshots and relevant information through an official, secure channel. I’m interested in what you’re building. I also need a workable support process when account access goes wrong. Please help me find a next step beyond the recovery links I have already exhausted.

@GeminiApp When we get gemini pro 4 model and waiting for benchmarks

@GeminiApp The ability to kick off research by voice and keep using your phone is a genuinely useful workflow improvement.

@GeminiApp Voice deep research sounds so convenient

@GeminiApp that used the foreground sync permission probably, I had the exact idea on my side pretty useful. if you want this feature to survive more robustly then cache the flutter engine while this runs.

@GeminiApp Why can't I request a video or image to be generated mid-conversation and instead I am required to start a new chat, losing the entire conversation context.

@GeminiApp Very interested in Gangstaling Topic.

@GeminiApp It would be cool to be able to do this using a frontier model.

@GeminiApp Nice, voice-first deep research is a useful combo. Does the report keep building in the background if you switch apps, or only after the prompt is sent?

@GeminiApp Bold launch day: every frontier CEO is on record saying "slow down," and Google ships voice-driven Deep Research anyway. The market noticed the difference — GOOGL +3.2% while SOXX bled -5.6%. Talk is cheap. Shipping isn't.

@GeminiApp is gemini the reason why i hijacked by crappy yahoo searches when using google ?

@GeminiApp

@GeminiApp google's finally winning the ai assistant war by letting apple take the reputational risk while they pocket the licensing fees.

@GeminiApp Now Gemini looks insane. abhi mazaaa aiyyyaaa. This feature not existed in any other ai providers. Good keep rocking bro.

@GeminiApp 👍

Hey Super, build a The application must feature an interactive voice-prompt simulator, a real-time Tone.js-powered audio waveform visualizer, a multi-agent synthesis progress ticker, an inspectable investigation node tree, and a structured report exporter. In the first viewport, visitors can select or type a complex research topic (e.g. quantum error correction scaling), adjust conversational depth, and click 'Start Deep Research' to trigger a simulated voice session. Tone.js animates a live frequency waveform while background worker agents synthesize sub-questions, crawl mock academic and web sources, and populate a live evidence graph. Users can click any sub-node to inspect confidence metrics, synthesized findings, and raw telemetry. Every action updates a canonical result schema that drives both the visual interface and the downloadable JSON/Markdown export.