it feels kind of stupid to learn an AI tool when the underlying model might become obsolete the next day so in this video I'm focusing on principles use cases and tips for Google Gemini that actually last let's get started diving right into a setting most people don't know about you actually need to go into your Gmail yes Gmail to get the most out of Google Gemini go into settings and under the general tab press command and control F and search for smart feature okay and you need to turn both of these on smart features enable this and Google Workspace smart features click this and toggle these on and click save you'll see how this works in a little bit next head on over to the actual Gemini web app and under settings go to apps and for me I only ena keep the Google Workspace and YouTube app/ extensions enabled because I don't want to waste time triggering these other apps by mistake since I never use them speaking of Gemini apps let's start off with my favorite use case visually it looks something like this step one share a YouTube video where an expert shares practical tips relevant for you step two ask Gemini to extract takeaways and frameworks from that YouTube video step three ask Gemini to apply the learnings to your own work for example using the add command in Google Gemini I'm going to bring up the YouTube app head on over to YouTube copy and paste Philip Hum's viral storytelling video i'm going to paste the link in enter extract the top three takeaways and explain them like I'm five this gives me actionable tips for example when telling a story I should zoom into one tiny exciting part because focusing on the small stuff makes a story feel real cool and it even shares a storytelling framework t AED thoughts actions emotions and dialogue next I can share a rough draft of a presentation from work asked Gemini to apply the TAD storytelling framework and share five actionable recommendations I can take immediately the reply is absolutely incredible for example this first recommendation instead of starting directly with OKRs and summaries which I originally planned to do let's begin with a core challenger or insight that drove this plan right and Gemini actually gives an example of a powerful opening for my presentation as for the other workspace apps I honestly don't use them that much because the agent capabilities aren't there yet for example although it's technically possible for Gemini to create a calendar Google calendar create a note and keep and search our Gmail it's much faster and easier for us to do that ourselves the one exception is using the mobile version of Google Gemini to create tasks due on specific dates using our voice which I do a lot when I'm out and about pro tip if you're using an Android phone you can ask Gemini to open up specific settings as opposed to wasting time trying to figure out where a page is buried next an extremely important topic for all AI users knowing when to use chat models and when to use reasoning models clicking into the drop down in the top left here you'll see that I intentionally blurred out the model numbers because they change too quickly right what's more important are the descriptions below for everyday tasks and best for complex tasks diving right to an example one of my pin chats in Gemini uses the everyday tasks flash model aka a basic chat model and this conversation thread converts simplified Chinese characters into traditional Chinese characters because I work in both languages the task is super simple i need the results fast and I don't need Gemini to think through any complications or nuances compare this to a situation where I receive an email in formal business Chinese and I need to translate this into English for my English- speakaking colleagues in this case I would ask the pro model aka a reasoning model to analyze the email in detail extract implications and share nuances English speakers might miss and although it's much much slower I'm more than happy to let Gemini take its time the latent space newsletter has a great visualization breaking this down chat models usually require several back and forths until we reach a good enough answer with reasoning models we provide the goal and the model itself plans the steps self-corrects if needed then gives a final response so my rule of thumb is to always default to the ProGemini model since the tasks I use AI for are generally more complex and I would only switch to Flash when the request is simple and or I want the answer really quickly here's a quick overview of the two types of models for some of the major players chat models for Gemini it's the Flash models for ChachiBT it's the GBT hyphen number models and for Deepseek it's the V models reasoning models for Gemini it's the Pro models for Chacht it's the O number models and for DeepSeek it's the R models at this point some of you might be wondering well Jeff when should I use the middle option the one that says uses advanced reasoning to massively oversimplify this is a hybrid model that is mainly geared towards developers who are looking for a balance between speed and cost so if you're just a normal user like myself don't worry about it default to the pro model and switch to Flash when you need faster response times by the way if you mainly use Google Workspace apps at work you might want to check out my Workspace Academy course where you'll learn a powerful workflow to eliminate mental and digital clutter i'll leave a link down below speaking of the pro model when I was preparing for my last video on AI agents I asked Gemini to turn this dense AI agent white paper into an interactive game to test my knowledge all I did was I uploaded the PDF onto Google Gemini selected Lea's Pro model enabled the canvas feature and asked Gemini to turn this paper into an interactive fun game designed to test my knowledge the question should start off easy and get progressively harder etc i will leave a link to this prompt down below if you want to try this yourself dem 9 will spend a minute or two writing the code running the code fixing its own errors if any and after all of that the end result is pretty mind-blowing the game is fully functional it looks pretty good and there are even elements like a working progress bar to be clear both Kajab BT and Claude have equivalents of the canvas feature but Gemini currently delivers the most complete result after just one prompt meaning you need to follow up a few times with Chachi and Claude to get something Gemini gets right the first time now what makes Google Gemini unique is not its underlying model everyone is overtaking everyone else every week right but it's deep integration within the Google Workspace ecosystem in the form of the Gemini side panel which can only be accessed after you enable the smart feature settings in Gmail speaking of let's say I receive an email from my colleague asking me to be a speaker at their offsite i want to say yes but I'm super busy right now i kind of want more context before I commit and I want to push a time back so as you can see my reply is not going to be a straightforward answer in situations like this it's just so much easier to open up the Gemini side panel use my Techuspanner app to bring up my email reply prompt input a bunch of unstructured thoughts and Gemini outputs a coherent and professional response I can just paste into my reply although the side panel is a paid Gemini feature if you're in the Google Workspace ecosystem it's definitely worth the decreased friction of accessing Gemini in the tools used every day for example in Google Sheets assuming I said yes to the workshop because I'm such a people pleaser I would ask Gemini to create a table that tracks the number of registrations based on the invitation channel instead of creating columns and rows from scratch this rough draft is a great starting point for me i can just remove the unnecessary columns and values make a few adjustments and it's good to go while we're in Sheets check this out here I've collected a bunch of feedback for this app and instead of using a traditional formula I can type equal AI open bracket quotation mark and start prompting for example I want to categorize all this feedback in column D under price UIUX privacy or others i'm going to type quotation mark again comma and I'm going to select this cell right here close bracket press enter i'm going to drag all those down command and control D and I'm just going to click this generate and insert and now Gemini will start categorizing the feedback from column D into those buckets that's pretty insane next and you can probably see where I'm going with this i can type equal AI open bracket quotation mark translate this feedback into simplified Chinese select this close the bracket let me just let's just give a few examples generate and insert basically if you don't remember the formula for something just describe the end result you want moving over to Google Docs let's say I'm preparing updates for the team look how convenient this is opening up the Gemini side panel i use at to find the Google sheet with the feedback and I type in a prompt received a lot of feedback write a short paragraph explaining what we're going to do press enter and let it run and with just one click I can insert the output into my Google doc document right and notice how all that was achieved without leaving this page pro tip for lengthy Google docs like meeting notes documents and product comms docs you can use the at command to insert an AI summary block up top that provides an at a glance overview of the document and whenever new content is added we can simply click the refresh button to get the latest changes and of course we can't forget about Google Drive here I like to rightclick on a project folder and ask for a summary of all the files within the folder first then I would ask Gemini to draft a detailed yet clear and concise report on the project I can use to debrief stakeholders and senior leadership to quickly recap Gemini's integration within Google Workspace is its biggest advantage since it helps us minimize context switching and fragmented workflows speaking of unfair advantages many of you might have heard about Gemini's massive context window but what exactly does that mean regular viewers know that I mainly use Claude for copywriting but look what happens when I try to add lengthy industry reports into a Claude project granted this document is 250 pages but it's already at 222% of capacity claude can't process this gemini on the other hand if I open up the settings for this gem which is the closest thing Gemini has to cla projects it's not even close to the upper limit of its context window even with three of those documents meaning for tasks where I need the AI to ingest a massive amount of data like analyzing industry trends Gemini currently dominates at this point I have to address what I believe to be Google Gemini's biggest downside how sensitive it is for context I graduated from Emory University and I know I know it's not in the top 20 in the US anymore and for the record the ranking only dropped after I graduated you know not that I care about the ranking i'm over it it's been 10 years i'm not insecure you are uh where was I um yes Google Gemini sensitive not me anyways former President Jimmy Carter would visit and give talks at Emery all the time and when he passed I saw in this article that Jimmy Carter was remembered more for his failures and because I was genuinely curious why that's the case I asked Gemini to explain and I had to try multiple times because Gemini would initially refuse to answer because the topic is too sensitive like what why and when I asked Chachib and Perplexity same question they were able to give me pretty objective and unbiased answers to be clear I totally understand the importance of AI safety principles but if too many legitimate requests start getting blocked I feel like I would just default to another AI app altogether but maybe that's just me let me know what you guys think if you enjoyed these tips check out my claw tutorial next see you on the next video in the meantime have a great one