3 Sources
[1]
Why I abandoned Gemini and returned to Google Assistant for my daily routines
I've been covering Android and other mobile technology for close to ten years now, with a specific interest in phone accessories, e-readers, and what makes each individual phone different from another. I delight in looking at the phone market from as many angles as possible, and while my opinions may be odd, at times, they're always from the heart as much as the head. I have a background in the mobile accessories world, which explains my odd enthusiasm for cases and things that clip onto smartphones. I worked for Digital Trends from 2017 to 2025. We live in a world packed full of AI, and it's becoming almost impossible to avoid using it. I've let myself use it more and more in the last few months, and I've had mixed results. For all Google's boasts, Gemini still feels like a toy, and while it's enjoyable in some contexts, it's just bad in others. Which might be why, now that Gemini is knocking on the door of every part of my Android experience, I've purposefully decided to go back in time. I've returned to using Google Assistant, and here's why. Gemini's compute limits mean I can no longer trust it as my assistant Google is pushing Gemini everywhere, but why should I use an assistant with limits? Posts 73 By Mark Jansen Sometimes a task needs doing, not thinking about I'm prone to procrastinating. Starting tasks is hard. Chopping it up into smaller chunks helps a lot, but I still struggle with it from time to time. And every now and again, I've got nothing on Gemini. Gemini loves to think. Ask it a question, and it'll spend the next thirty seconds or so umming and ahhing, going over the alternatives, and eventually delivering you an answer. Still, it's not really thinking -- it's checking a database and running probability mathematics. But the outcome is the same. It takes too long for Gemini to get to the end of its train of thought. All the while, I'm waiting for it and getting more and more annoyed. To make things worse, Gemini is so chatty. It loves to talk, and it'll waffle on about something that's, realistically, only tangentially related to what I've asked it. Gemini will happily go on and on about the history of light switches, after it's taken its sweet time gathering that information. I'm happy for you, Gem, but what I actually need is someone to turn the lights on. Google Assistant just does it. And most of the time, that's all I want. It's hard to trust Gemini I can't deny that Gemini is a useful piece of technology. For some tasks, it's hard to beat it or any of its loquacious kin. Sometimes, when you need an answer, being able to quickly ask a chatbot out loud is a very useful tool, and I appreciate that. But how much can I really, actually trust Gemini? I was on a long drive this weekend, and I knew I needed a rest stop within the next half an hour or so. I knew roughly where one was, but I wanted to make sure I didn't overshoot it. So I booted up Gemini through Android Auto and asked it how far away it was. The answer came in, and it was about what I expected. And then Gemini asked me if I wanted to add it as my next stop in my Google Maps navigation. I paused. It would make sure I wouldn't miss it. But something stopped me from confirming. It was a niggling doubt. I realized at that moment that I don't trust Gemini all that much. It's not because I thought it might be lying to me. No, it was because I didn't know whether the well-known chain restaurant I'd asked it to find was the correct one. After all, it might have decided to take me into some nearby town center to one there, and not the one on my route. I was close enough to a town for that to be the case, and because I was driving, I couldn't check to be sure it was right. So I told it, "No, I'd find it myself." Gemini didn't comment, but if it had been a human, I wondered if it would have noticed my hesitance. My clear and obvious lack of trust in it. In fairness to Gemini, it can do this, and Google Assistant can't. Gemini can take my request, cross-reference it with the information on the map, my GPS location, and the traffic on the road, and make a judgment on how long it'll take me to get there. Google Assistant just can't do that. It can read out web pages, but it's pretty bad at it, so most people don't use it at all. But is it better to have a feature and not trust it, or just not have the feature at all? I'm starting to think I'd be happier without bad or faulty options. Gemini isn't really a replacement for Google Assistant Recent months have shown that Google intends to replace Google Search with a Gemini-powered AI search function. There are many problems with that, which I won't get into here, but I will say it shows exactly where Google's head was at when it was designing Gemini. When you see Gemini as a Google Search replacement, rather than a Google Assistant successor, it makes a lot more sense. It's why it lacks a number of the abilities Google Assistant has, like native support for Routines, and being able to directly control other devices. That's a big part of why I'm switching back to Google Assistant for some of my devices, like my remaining Google Nest speakers. While Google Assistant may not be able to hold a conversation or give me a long answer about what the different types of spring are, it can actually do what I want it to. Google Assistant is the better option if you want a voice assistant that can do stuff. Because while it may have a more limited toolset, it actually knows how to use them. That makes a huge difference.
[2]
Gemini on Android needs a "dumb" mode (because it's too smart for its own good)
Gemini regularly impresses me by carrying out complex, multifaceted requests. That's why it's so infuriating when it fails spectacularly at the most basic tasks -- things Google Assistant could handle with ease. There's a simple solution to this problem: make Gemini dumber. Many Android fans long for the days of Google Assistant (sometimes I'm one of them). Sure, it wasn't as "intelligent" as Gemini, but it was excellent at following simple commands. Ironically, that's where Gemini struggles the most -- as a digital assistant. Why Google Assistant was better at simple tasks A smaller toolbox to work with The main issue with Gemini as a mobile assistant is how it handles commands. As a Large Language Model (LLM), it processes every prompt it's given and generates a response. This is why you may have noticed Gemini will respond slightly differently almost every time to the exact same requests. Google Assistant, on the other hand, was built to work with direct commands and phrases. When it recognized a command, it would match it to the associated action -- like a switchboard operator. This allowed Assistant to respond quickly and accurately if it understood the task. Let's use the command "Set a timer for 5 minutes" as an example. Gemini reads the text and attempts to conceptually "understand" the intent: starting a countdown. It then has to recognize that it can't carry out that task on its own, and to choose the appropriate extension to do it. If all of that goes smoothly, Gemini tells the clock app to start a timer. Google Assistant simply scanned the text for specific intent phrases. In this case, it would match "set a timer" to the action of creating a timer, and "5 minutes" to the duration length. Without any "thinking," the command was sent directly to the clock app. Google Home Speaker $99 at Google Store Expand Collapse Some tasks don't require "thinking" It's not that deep The benefit of Gemini's approach is that it can understand less specific wording and do more with it. I've put this to the test by saying, "Set a timer for honeybush blueberry pancake tea." Gemini looked up the variety of tea, found the recommended steep time, and started a timer for that length. I was impressed. If I gave that same command to Google Assistant, it would recognize the "Set a timer" part, but since I didn't mention a time, it would have to ask how long the timer should be. The problem with Gemini's high-level approach is that it's just not needed for many of the digital-assistant-type tasks I try to do on my phone and smart speakers. A recent conversation with Gemini really hammered this point home. As a Google product, Gemini has access to my Google services. I use both Keep and Tasks for different things -- Tasks is where I have my primary "to-do" list. However, Gemini always defaults to using Keep when I ask to add something to my to-do list. Unlike Assistant, I can't simply choose which list app I want Gemini to use. So, I either need to specify to use Google Tasks every time, which is tedious, or I can add it to my personal instructions. I did the latter: If I ask you to add something on my to-do list, put it on my Google Tasks list called "To-Do". Do not use Google Keep. Now, I know for a fact that Gemini can access my Google Tasks -- it's done it many times. That's why I was very annoyed when I recently asked it to add something to my to-do list, and the conversation below proceeded to happen. TL;DR: Gemini said it couldn't access Google Tasks. I pointed out that I had just used Tasks in the previous conversation. Gemini doubled down. We went back and forth. Finally, I asked what was on my Tasks list, and -- surprise! -- Gemini could tell me. It apologized for the "confusion." Gemini needs a "dumb" fallback There are times like the above when Gemini is too "smart" for its own good. There are so many common, daily life tasks that not only don't require an LLM, but are actively worse when using one. It's like watering a house plant with a pressure washer. Deals Save on Phones & Mobile Deals: Top discounts now Find limited-time deals and meaningful savings on phones, mobile accessories, and companion gear. Shop discounted handsets, chargers, earbuds, cases, and smart speaker bundles. Explore offers to compare discounts, catch price drops, and add savings on accessories and protection. Deals Explore Phones & Mobile Deals The perfect solution would be some sort of "dumb" mode that allowed Gemini to function more like Google Assistant in certain situations. It's a bit ridiculous to be using gallons of water to set timers, create reminders, turn on lights, play music, and get the weather forecast when it was perfectly possible to do all of that before AI. Gemini's full abilities are awesome for some things -- genuinely -- but a little restraint would go a long way when it comes to being an assistant on my phone. We should be able to have the best of both worlds. Gemini is failing at a basic Android task my phone nailed back in 2014 We've been told over and over again how Gemini is an "upgrade" over Google Assistant. In many ways, it is -- but the areas where it still lags behind feel infuriating. Google says Gemini can help you remember where you parked, but the reality feels embarrassing. Posts 4 By Joe Fedewa
[3]
7 hidden Gemini assistant features that organized my chaotic work tasks
Jade Bryan is a Feature Author at Android Police with over a decade of tech reviews and news reporting experience, stretching back to the Symbian era and the early days of Android. He covers the breadth of the mobile industry, specializing in operating system deep-dives, app testing, and hardware analysis. In his spare time, you can find him tinkering with his Android devices and wearables, seeking ways tech can improve daily life. He also enjoys playing MMORPGs, traveling, or feeding stray cats and dogs. When Google launched Gemini, I wasn't exactly thrilled. Although I have used Google Assistant before, I don't consider myself an AI fanboy by any means. I actively avoided it on my personal devices for as long as possible. But since I write tech for a living, I can't ignore it. Eventually, my job forced me to give it a fair try. I started small by experimenting with standard features like letting it summarize long YouTube video formats and messing around with screen sharing, which, to my delight, immediately hooked me. Having contextual and tailored answers proved to be more useful than I initially thought. Eventually, I caved. Since then, I've been taking advantage of Gemini in juggling every day and work tasks. I use it regularly to triage multi-threaded emails in Gmail and to fly through documents across Google Keep, Calendar, and Sheets, among others. It has evolved from an annoying gimmick into a utilitarian digital assistant that keeps my daily workflow running smoothly. If your daily routine feels like a constant mess, here are the exact Gemini features that sorted my chaotic tasks. I used Gemini wrong for months, here's the setup that actually works Stop using Gemini like ChatGPT Posts 11 By Parth Shah Create voice notes on the lock screen Record thoughts without going through biometrics I prefer keeping my phone securely locked down at all times. But it is incredibly annoying to constantly fumble with the fingerprint scanner or selfie camera to log an immediate note because I stumbled upon a great source or had a fleeting thought. My short-term memory is quite unreliable, so I capture random ideas the second they come in. One of the features that helped me organize and avoid losing those thoughts is the ability to create voice notes without unlocking my Google Pixel 9 Pro XL or Samsung Galaxy Tab S10 FE. Google finally brought this system-level capability to Gemini, expanding its reach to connected Workspace extensions like Keep and Tasks. Setting it up is straightforward. Launch Gemini and then open Gemini Settings, navigate to the Gemini on lock screen, and toggle on the feature. When it's active, you can start dictating thoughts on the fly without unlocking your device. It is significantly ahead of the old Google Assistant implementation because Gemini understands natural, conversational language instead of choking on rigid commands. My only real gripe is that I wish Google would let us edit existing notes or tasks directly from the lock screen. Currently, when you need to adjust or tweak data in Google Keep or Tasks, you have to unlock your phone. Even with that limitation, this feature alone has saved a lot of time and reclaimed a ton of my best ideas from disappearing into thin air. Create recurring tasks using Gemini And save a lot of time As a tech writer, a massive chunk of my day revolves around scouring the internet for news and the best deals. Doing that manually every morning is tedious, but Gemini automates the routine. By leveraging the chatbot, I can create highly specific, recurring background tasks that do the heavy lifting for me. For example, I set up a custom Gem to hunt for discounts on the Pixel 10 Pro. I fed it my preferred retailers, my variant choices, and a rule to ignore any price cut under $100. It runs this like clockwork every morning, and I programmed it to skip weekends and holidays. Beyond news and deals, I can leverage it for other categories and topics. It is a flexible capability that has saved me from wasting hours going through each retail store. You can set up basic automated prompts through standard Gemini chat or in the Scheduled tasks in the settings. Alternatively, you can use Gemini Gems to add specific rules and nuances. You can create a scheduled task on Gemini without going through Scheduled actions in the settings or Gems. Ask Gemini and add to Calendar A smarter Google Lens alternative I have been a die-hard Google Lens power user for ages. But when Gemini's screen-sharing integration and Ask Gemini capabilities rolled out, I decided to test it to see if it could be a viable alternative. After putting it under my belt, it completely convinced me to drop Lens. Gemini offers several advantages, such as handling complex visual data and reducing the number of steps in a standard visual search. For everything related to scheduling and meetings, I keep track of most of them in Calendar, but everything used to be a massive chore. Now, if I am looking at a chaotic schedule inside an email or image, I pull up the Ask Gemini overlay on my Pixel. It extracts the relevant dates and times, without me having to copy and paste a single line of text. From there, I tell it to create the event straight into my calendar, which it handles convincingly. More importantly, I can surface my upcoming agenda from the lock screen. With a quick voice prompt, I can command it to remind me hours or days before a meeting. Connect YouTube summaries to notes and maps Extract data using only a video URL YouTube summarization was one of Gemini's earliest tricks, and it remains one of my favorites. I watch a considerable amount of game and hardware reviews and comparisons before planning to make a purchase. I also rely heavily on travel vlogs to map our multi-day family vacations. I'm not sure if it is the same for everyone, but sitting through a 25-minute DIY video to get a few steps or travel recommendations feels exhausting. Instead, I feed Gemini the YouTube URL and let it extract and highlight details or create bullet points. It's a massive time-saver, but there's one more useful feature to that. Because of the integration with other apps, I can copy that extracted data into a Keep note or Docs, or automatically add stops in Google Maps for places and attractions. Manage Google Sheets via commands The smart autofill is a lifesaver Over the years, I've come to realize how invaluable Google Sheets is for my daily workload. It has become a valuable tool for managing my tasks, such as real-time headline aggregation and mapping travel itineraries. While the mobile app is handy for quick edits on the go, I didn't truly appreciate its potential until Google integrated Gemini into the web version of Sheets (which you can access through the Google One Pro plan). For my daily writing workload, I usually dump a massive, unorganized list of potential article topics into a column. I have Gemini analyze and rank these headlines by priority. Even better, it can cross-reference the live web to instantly check which of those topics I have covered or published elsewhere. It also works well for managing dates and locations. I can type a command like: Help me organize our 7-day itinerary to Taipei using these places and flight details. Gemini easily adapts to specific nuances, too, like requesting a light first day or arrival to manage travel fatigue or suggesting a strictly kid-friendly route. Besides the integration, Gemini's smart autofill feature is easily my favorite tool in the entire suite. It's a nice alternative to the classic formulas. The autofill feature analyzes the surrounding context of your completed rows or columns and intelligently populates blank cells. This saves me a lot of time figuring out the formula while avoiding breaking the sheets. The Google Sheets 'miracle' that finally organized my contacts One Gemini prompt to clean up years of mess Posts By Rajesh Pandey Do tasks right in Gemini chat You don't need to leave the chatbot If I am deep in a Gemini thread, the last thing I want to do is navigate away and launch a separate app just for one action. It's pure unproductive. Google solved this friction by embedding its services directly into Gemini's chat box. All you do is type the @ symbol followed by the name of the app you need to summon, such as @Gmail or @Calendar, and Gemini searches for it on the spot. It works flawlessly with voice commands, too, provided you clearly mention which app or service you want to target. For instance, I use it to audit my morning Gmail inbox. I can tell Gemini to scan my unread emails and surface only the critical entries. It effectively filters out promotional spam and irrelevant press releases. It also works when you're tracking a package, as it can surface my inbox, copy the tracking number, and do the tracking for me without leaving the chatbot. However, there are minor shortcomings, such as it not getting all the details, as the provider may limit the data available. Meanwhile, in Google Docs, I use Gemini to track documents and keywords without wading through a stockpile of articles and drafts. Access smart home devices from the lock screen Turn a bulb in Gemini on or off Beyond using app-based features to keep my tasks in a tab, I've found that Gemini is a practical assistant for Google Home. It's handy for small tasks, like spinning up my Xiaomi air purifier a few minutes before I walk into the room, or increasing the brightness of my desk lamp at sunset. Even better, it works directly on the lock screen and includes a one-tap widget that shows my smart devices' status. It isn't as groundbreaking as other Gemini productivity capabilities. But it's a minor convenience for eliminating a bottleneck in my workflow.
Share
Copy Link
Google's aggressive push to replace Google Assistant with its Gemini AI assistant is creating a divide among Android users. While some embrace Gemini's advanced capabilities for task organization, others are abandoning it due to frustrations with overcomplicated responses to simple commands. The debate highlights a fundamental tension between sophisticated AI and practical usability in consumer tech.
Google's transition from Google Assistant to its Gemini AI assistant is proving divisive among Android users, with mounting criticism over the digital assistant's handling of basic tasks. While Google aggressively integrates Gemini on Android across its ecosystem, users report significant frustrations with Gemini vs Google Assistant performance, particularly for daily routines and tasks that previously worked seamlessly
1
2
.Source: How-To Geek
The core issue stems from how the Gemini AI assistant processes commands. As a Large Language Model, Gemini analyzes every request conceptually, generating slightly different responses each time. For a simple command like "Set a timer for 5 minutes," LLM processing requires Gemini to understand the intent, recognize it cannot perform the task independently, and select the appropriate extension to execute it. Google Assistant, by contrast, matched specific intent phrases directly to actions, enabling faster and more reliable execution
2
.User trust has emerged as a critical concern with the new AI assistant. One user recounted a troubling experience during a road trip when Gemini offered to add a rest stop to Google Maps navigation through Android Auto. Despite receiving what appeared to be accurate information, the user hesitated, uncertain whether Gemini had identified the correct location along the route or a different one in a nearby town. This lack of confidence in contextual responses led the user to reject Gemini's assistance entirely
1
.The practical usability in consumer tech suffers further from Gemini's verbose nature. Users report the digital assistant takes excessive time processing requests, then delivers lengthy explanations for tasks requiring simple execution. "Gemini will happily go on and on about the history of light switches, after it's taken its sweet time gathering that information," one frustrated user noted, adding that they simply needed the lights turned on
1
.Critics argue that Gemini functions better as a Google Search replacement than a Google Assistant successor, explaining why it lacks native support for many features Assistant handled effortlessly. The voice commands experience particularly suffers, with users requesting a "dumb mode" that would allow Gemini to bypass LLM processing for routine tasks like setting timers, creating reminders, controlling smart features, and checking weather
2
.Source: Android Police
One user experienced repeated failures when asking Gemini to add items to Google Tasks instead of Google Keep. Despite setting explicit personal instructions directing Gemini to use Tasks, the AI assistant claimed it couldn't access the service. Only after multiple exchanges did Gemini successfully retrieve the Tasks list, apologizing for the "confusion"
2
.Related Stories
Despite widespread user frustrations with Gemini, some Android users report significant productivity gains after discovering lesser-known features. Integration with Google Workspace enables sophisticated task organization capabilities, including lock screen voice notes that sync with Keep and Tasks without requiring biometric authentication. This feature captures fleeting thoughts immediately, addressing a common pain point with the previous Assistant implementation
3
.
Source: Android Police
Recurring automated tasks represent another area where Gemini's advanced capabilities shine. Users can create custom Gems that execute specific searches on schedules, such as monitoring retailer websites for product discounts above certain thresholds while excluding weekends and holidays. The Ask Gemini screen-sharing feature has also convinced some Google Lens power users to switch, offering advantages in handling complex visual data and reducing steps in visual searches
3
.The split reaction to Gemini highlights a fundamental question facing AI development: whether sophisticated intelligence that occasionally fails at simple tasks proves more valuable than reliable execution of basic commands. As Google continues pushing Gemini across Android devices, the company faces pressure to address practical usability concerns while preserving the advanced capabilities that some users find transformative for daily routines.🟡 familiarity with it.
Summarized by
Navi
[3]
27 May 2026•Technology

13 Jun 2026•Technology

07 Jul 2026•Technology

1
Technology

2
Policy and Regulation

3
Business and Economy
