Everyday AI Podcast – An AI and ChatGPT Podcast - EP 599: 5 New Overlooked ChatGPT Features You Should Be Using But Aren’t

Episode Date: August 28, 2025

The GPT-5 rollout was messy. Then, Google went AI ship crazy. In between all of that, OpenAI released some powerhouse features inside ChatGPT that seemingly no one is paying attention to. Join us a...s we uncover them and give you a leg up on everyone else. Newsletter: Sign up for our free daily newsletterMore on this Episode: Episode PageJoin the discussion: Thoughts on this? Join the convo and connect with other AI leaders on LinkedIn.Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineupWebsite: YourEverydayAI.comEmail The Show: info@youreverydayai.comConnect with Jordan on LinkedInTopics Covered in This Episode:Five Overlooked ChatGPT Features RecapFlashcards with Quiz GPT Interactive ToolCustom ChatGPT Personalities ExplainedAdvanced Voice Mode in ChatGPT & GPTsGmail and Google Calendar Auto ConnectorsCustom Instructions for ChatGPT ProjectsProject Folders vs. Custom GPT OrganizationChatGPT Agent Mode New Use CasesTimestamps:00:00 Overlooked ChatGPT 5 Features05:06 Unannounced OpenAI Updates Discussion09:02 Personalized Learning with LLMs10:58 GPT-4 Personalities Address Sycophancy14:12 Custom Instructions and Personalities in ChatGPT18:00 Custom GPT Voice Limitations21:38 Streamlining Email with AI Prompts25:10 Custom Chat Instructions Toggle28:43 Customizing ChatGPT: Flexibility Challenges31:02 "AI Updates and Sharing Instructions"Keywords:ChatGPT, GPT-5, GPT 4o, OpenAI, ChatGPT features, overlooked ChatGPT updates, custom personalities, flashcards, GPT quiz, interactive quiz, advanced voice mode, voice mode updates, ChatGPT connectors, Gmail connector, Google Calendar connector, auto connectors, custom instructions, ChatGPT projects, project memory, ChatGPT organization, ChatGPT folders, project only memory, memory settings, ChatGPT system prompt, ChatGPT hallucinations, ChatGPT prompts, ChatGPT deep research, custom GPTs, Canvas mode, Notebook LM, Gemini, Gemini live, Claude, Anthropic Claude, email management with AI, AI productivity tools, AI for business leaders, AI learning tools, AI-powered flashcards, interactive learning AI, personalized AI, AI chat modes, sycophantic GPT, ChatGPT tone settings, ChatGPT settings, AI updates 2025, AI task automation, AI-driven workflow, ChatGPT troubleshootingSend Everyday AI and Jordan a text message. (We can't reply back unless you leave contact info) Start Here ▶️Not sure where to start when it comes to AI? Start with our Start Here Series. You can listen to the first drop -- Episode 691 -- or get free access to our Inner Cricle community and all episodes: StartHereSeries.com Also, here's a link to the entire series on a Spotify playlist. 

Transcript
Discussion (0)
Starting point is 00:00:00 This is the Everyday AI Show, the everyday podcast where we simplify AI and bring its power to your fingertips. Listen daily for practical advice to boost your career, business, and everyday life. Meet Firefly AI Assistant, now live in Adobe Firefly, the all-in-one creative AI studio. Just describe what you want to create and the assistant handles the rest, orchestrating multi-step workflows across Photoshop, Premiere Express, and more in one conversational interface. You direct the outcome. The assistant accelerates execution. Even if you're a power user of chat GPT,
Starting point is 00:00:49 there's been a dozen or so extremely useful and powerful features that OpenAI has added over the last few weeks that I think almost everyone missed. That's because they came in right around before, during or after the rocky rollout of GPT5, and Google's been absolutely dominating the headlines ever since the GPT5 release. So I've been kind of sitting around like, I wonder when people are going to start talking about some of these low-key overlooked features and people still aren't. So that's what we're going to be going over today. I'm going to be talking about five new overlooked chat GPT features you should be using, but you probably aren't. All right, I'm excited
Starting point is 00:01:38 for today's episode. I hope you are too. Let's get started. What's going on, y'all? My name's Jordan Wilson and welcome to Everyday AI. This is your daily live stream podcast and free daily newsletter that's helping everyday business leaders like you and me, not just keep up with what's happening in the world of AI, but how we can use all the info to get ahead to grow our companies and our careers. It's unedited, unscripted, bringing you the realist thing in artificial intelligence. It starts here. But if you want to take it to the next level, that's going to be on our website, your everyday AI.com. Go sign up for the free daily newsletter where we're going to be recapping the highlights from today's show. So if you miss something, or if you're like, oh, how do I do that?
Starting point is 00:02:17 I need that link. It's all going to be in the newsletter as well as all of the important AI news and updates for today. All right. You probably miss these features. So kind of out of nowhere, there was this huge backlash right after OpenAI released GPT5, right? There were some reasons for it. There was some things in the new model that just weren't working. There were some things that Open AI just straight up screwed up, right? Even some slides on their presentation, right? People always give me a hard time when I misspellings on mine.
Starting point is 00:02:53 I'm like, look at OpenAI screwed up their presentation as well. And then they got rid of all of these. old models that people really cared about and they relied on for their day-to-day use, this is GPT-40. So there was such this huge backlash a couple of weeks ago when opening I released GPT5 that there was all these actually killer features that no one is still talking about because right after that, Google went straight up silly and just seemingly day after day was releasing new and powerful AI updates.
Starting point is 00:03:23 So today, we're going to go over these things. We're going to go over those five new and overlooked features of chat chbt. I'm going to talk about what they are and why I think they've been slept on. And I'm going to give you practical and actionable ways you can use these features. All right. Let's get straight to it. Live stream audience, let me know how many of these features out of five. Just drop something in.
Starting point is 00:03:52 I'm always curious. And podcast audience, if you ever want to join the live stream, we do this every day, 7.30 a.m. Central standard time. So here's the five features. I'm not going to make you wait, but live stream audience, give me a number zero to five. How many of these are you actually using? So flashcards with GPT quiz,
Starting point is 00:04:13 it's just like a hidden feature that allows you to create flashcards. Really cool. I'm going to show you how. Custom personalities. It was kind of mentioned in the GPT5 live stream, but no one really talked about it or has visited it again. Then you have the advanced voice mode access updates. So a new advanced voice mode. And now advanced voice mode is available in GPTs, which I think most people are talking about and don't know.
Starting point is 00:04:44 Then you have Gmail and Google Calendar on autopilot as connectors. So previously, these had to be deep research only, which wasn't really useful. But now you can get in. instant updates from your Gmail and Google Calendar once you've connected it. And then one thing I've been asking people at OpenAI for for a long time is custom instructions for projects. So there you go. Five flashcards. Four, custom personalities. Three, advanced voice mode access and updates in GPT.
Starting point is 00:05:19 Two, Gmail and Google Calendar on autopilot. And then number one, custom instruction updates. All right. Let's be honest. All right. Mark, Mark says two. Brian says two. Jackie says three.
Starting point is 00:05:35 That's the thing. And I know, hey, a lot of these people that just answered are some of our live stream regulars. So, yeah, even I think power users didn't number one know that these updates were announced. Some of them, Open AI, hasn't even publicly talked about and didn't even have any releases on. So, again, I think it's just the combination of. of timing, you know, having some of these things come out right before or after GPT5,
Starting point is 00:06:03 or maybe they were a part of GPT5 rollout. And then just the backlash after and everything Google's done. I think that we need to spend a little bit of time today talking about these. And actually, I hate when I do this. When I put together a list, I have my notes ready. And then I'm like, oh, crap, there is actually some way better ones that I didn't include. So literally four minutes before I started this live stream, I finish putting together a guide. So this is seven less known chatybd features because there was two I forgot.
Starting point is 00:06:37 And I'm like, oh, crap. So if you want access to those and they're on agents and I've literally these two new overlooked agent features are the agent features I'm using the most. And they were not out or available when Open AI first released agents. So if you want access to those other two and an entire kind of mini website with AI built into it, so you can talk to essentially this episode. Just share this and repost this on LinkedIn. And I'll share this with you. All right.
Starting point is 00:07:09 So let's start with number five. And that is our flashcards. So I kind of have an example of this for our live stream audience, but I'll kind of do my best to describe this. So all I saw on this on Open AIs, website was a small little footnote, I think, on a dev page, right? So this is weird. You don't have to be in the new chat GPT learning mode to use this. A lot of people thought they were connected. They're not. And it does seem a little finicky, but I'm going to give you some examples of exactly how you can
Starting point is 00:07:43 use this, but it makes essentially an interactive canvas. It looks kind of the same every single time. And then depending on what you want it to do, or you can give it information ahead of time. it's going to create a little interactive multiple choice quiz. So very cool. So in this example that I'm showing, and I do think you have to be on thinking mode for this to work. I've tried it on auto and fast and haven't got it to consistently pull up, but when using chat GPT's thinking mode, this does work.
Starting point is 00:08:15 So in this example, I said create a quiz with quiz GPT, one word, all right, and flashcards to create a quiz. on large language models in English, right? And then it creates an interactive quiz. Then I can go through and interact with. So this is really, really handy. So also, this is a big learning benefit that notebook L.M doesn't have yet.
Starting point is 00:08:40 This is actually, I follow all this stuff, obviously, because I'm a dork. But this is actually one of the most requested features for notebook LM. So if you don't know notebook LM, it's one of my favorite tools. I think it's one of the internet's most favorite tools. So, you know, it's essentially a grounded version of Google's Gemini model where you can just upload your information. And it's a great way. And it's my favorite way to learn because you can create custom video overviews,
Starting point is 00:09:07 you can create custom podcast overviews, mind maps. But the one thing missing out of notebook LM is just interactive flashcards. So this is something now you have inside of chat chit. So like I said, it is an in-chat quiz or flash card. And if you kind of call quiz GPT and tell it to quiz you on a topic and then say flashcards, it'll usually work. I don't know why it's not like a toggle or a button that you push. I don't know if this is kind of like a secret feature or what because Open AI hasn't really
Starting point is 00:09:37 talked about it a whole lot. So why should you use it? Well, I think a lot of people are going to assume that this is for students. And yeah, it's going to be helpful for students. But y'all, one thing that I cannot. overemphasized, and I've said it many times on the everyday AI show, is I think sometimes people look at large language models, you know, such as, you know, GPT5 inside of ChetGPT as a way to spend less time on a certain deliverable or on a certain output. I'd say that's the majority of the way
Starting point is 00:10:09 people use in a primary use case of a large language model. That's not mine. I use large language models to learn things at a better, in a better way and way more highly personalized, right? So obviously, you get to combine all the different powers and capabilities of chat GPT to first of all pull in this information. So you don't just have to open-endedly say, you know, hey, go create me a quiz on blank. You can share documents, you know, have a couple of back-and-forth conversations first with chat, GBT, and then go ahead and call out this quiz. GPT and say, hey, quiz me on all of this information above. You can tell it right now, you know, hey, I'm a beginner.
Starting point is 00:10:52 I'm intermediate. And you can also kind of give it some conditional rules for the quiz as well. So I think that's something that's really helpful. So yeah, has anyone else used the flashcards? They're really cool. I like them. All right. So that's number five.
Starting point is 00:11:06 Let's go to number four. Chat GBT. Personality. All right. So again, this was released with GPT. But it got like a little one minute mention in Open AI's like two hour live stream. And then I don't think a lot of people focused on it after that. And I think I could be wrong.
Starting point is 00:11:33 Open AI didn't say this, but I think these personalities were actually ways to address some of the mounting pressures of the sycophantic GPT4.0, right? So there was a period. They corrected it a little bit. But GPT40 essentially got to the point where it's just agreeing with you, right? Regardless of what you say, it's just your hype man. And it's not, it wasn't always helpful. Right.
Starting point is 00:11:58 Even my wife the other day said, you know, hey, it seems like chat GPT sometimes just agrees with whatever you're saying, right? So I think these personalities were almost meant to address some of those issues. So on chat GPT's website just updated a couple of weeks ago, they said, a personality is the style and tone chat GPT users use uses when responding to you. It's a combination of traits, voice, and behavior that shapes how answers feel, whether that is friendly and casual, concise and professional, or something else entirely. So it says changing the personality does not affect what chat GPT can or cannot do or the
Starting point is 00:12:39 rules it follows for safety. All right. So here's the different personality. that were released. So it's pretty simple. All you have to do is go into customize the customize chat GBT option inside of your chat GPD settings. It's kind of confusing because you don't go to the settings.
Starting point is 00:13:02 You click on kind of your icon. If you're using it on the web in the lower left hand corner and there's a little area that says customize chat GPD. And it's actually a really thin little thing. It just says what personality should chat. he has, it's always by default. It's the default, which is none, but then you can choose the others. So, cynic is dry, sarcastic, and blunt with witty asides. It's good for users who enjoy candid, no nonsense, guidance, and creative brainstorming with edge. Then you have robot, which is
Starting point is 00:13:34 precise, efficient, and emotionally flat with direct answers first. Ideal for technical tasks, checklist, troubleshooting, and fast, unambiguous instructions. Then you have listener. So, if you're missing the, you know, warmness and the back and forth of GPT4O, maybe go try listener. So this is warm, calm, and reflective with gentle, clarifying questions. Great for decision support, planning, and when you want a thoughtful sounding board. Then you have nerd, playful, curious, and explanation forward with context and mechanisms. Best for learning, deep dives, creative problem solving, and guided exploration. So if you've been using the new GPT5, right, whether you're kind of regardless of what mode you're using, and yes, Open AI keeps changing the mode names.
Starting point is 00:14:25 So now you have auto, instant, thinking MIDI, thinking, and pro. All right. So sometimes you're like, this is way too wordy or this is too short. Right. So these are great different personality modes that you can put in there. And then the good thing is, even if you choose one of these four custom personalities, you can still have your custom instructions. So they can still work hand in hand.
Starting point is 00:14:53 It's not like using one of these four personalities will then go ahead and override whatever custom instructions you use inside chat chit. And if you are brand new or if you're not sure what that means, essentially custom instructions inside chat chabit are kind of like a system prop. So it's a set of instructions that you apply to your entire set of chats, regardless of where those happen. So like as an example, generally I don't use custom instructions and I never have, at least for the way that most people are using them.
Starting point is 00:15:30 The only custom instructions I've ever used are essentially a set of like zero fabrication instructions. So try to cut down on hallucinations. to keep the answers clean, and then to also force with every single answer, to force chat GPT to give me a confidence score and a reason for it, because a lot of stuff, a lot of my day-to-day task that I'm doing revolve, sorry, involve a lot of kind of mini-rag involves a lot of, you know, information retrieval from the web. So I like to always have that set of custom instructions. But aside from that, I don't really and haven't really used custom instructions. But again, this chat chavit personalities is a great one.
Starting point is 00:16:16 And yeah, I don't know if anyone wants that set of custom instructions, happy to send it to you. If you just leave a comment, just say, you know, zero. And I can send you those just so I know. All right, let's keep going. Number three. And actually, no, let me know. Live stream minds. Have any of you used cynic robot listener or nerd?
Starting point is 00:16:39 I've toyed around with all of them except listener, at least for me, I don't want anything like that. I don't need warm, calm, reflective, asking me questions. That's usually not what I'm using chat chbtee for. So yeah, let me know. Have you used any of those? All right. So number three, there's some advanced voice mode updates.
Starting point is 00:16:58 And you can use now advanced voice mode inside of chat chpt. So this is from the open AI website. They said today we're rolling out improvements to voice mode to make it more accessible and useful for everyone. It also now works with custom GPTs. We're expanding access with near unlimited use for plus users. So if you're on the $20 a month plus plan and hours each day for free users. So yeah, you can, even if you are a free chat GPT user, you now have access to advanced
Starting point is 00:17:29 voice mode. So previously, it was kind of tiered. There was the standard voice mode, which was more neural. it wasn't a very human-esque cadence. It was very robotic. Whereas advanced voice mode, it is much more neural. It does, you know, it's faster. It sounds a lot more like talking with a human.
Starting point is 00:17:50 And advanced voice mode would, you know, it had those kind of noises in between the time you talk to it and its response. You know, so it was sometimes a two, three, four second delay, depending on what information you needed. Whereas advanced voice mode, it's almost instant. However, I will say this. I haven't been super impressed with the updates to advanced voice mode. I know a lot of other people have.
Starting point is 00:18:13 Maybe that's because I'm using a lot of Gemini Live. So there's just certain instances where I think their version, Google Gemini's version, is better. And also, it's important to know that in a couple of weeks, standard voice mode is going away. So like I said, the big updates here, in general, improved cadence and responses. It's available to free users. The standard mode is retiring in September. But let's talk about the small footnote here, which is it's now available inside of GPTs,
Starting point is 00:18:45 which is extremely important. However, I was doing a lot of testing on the voice mode inside of custom GPTs, and it doesn't work how you might want it to. That's because aside from web retrieval, most of GPT-5s, custom tooling, so all the fun scaffolding doesn't work. So, you know, as an example, you know, I was testing this out last night to confirm.
Starting point is 00:19:13 You know, so I just created a quick GPT, dropped an Excel sheet in there. So when I chat with it, it obviously pulls everything out of the Excel sheet perfectly. It can use canvas mode, all these other things. But when using voice mode within a custom GPT, it cannot access your files. The downside is it doesn't know that, and it will hallucinate hard. This is one of the most blatant hallucination kind of problems that Open AI probably, I would assume they don't know about it. Otherwise, they should have fixed it and patched it. Because right now, even if you do have project files uploaded inside of a custom GPT, they obviously work when you're doing text-to-text.
Starting point is 00:19:55 But if you're doing the advanced voice mode inside a custom GPT with project files, it's going to pretend it's pulling all the information. it's not. So keep that in mind. I hope they work. I hope they work that out. But right now, that's a big downfall. But however, I do still think using advanced voice mode in GVTs is extremely useful. It just can't be one of those things where you are hoping to get information from an uploaded file. All right. Adobe just introduced an entirely new way to create, bringing the power and precision of its creative suite into one conversational experience. Meet Firefly, AI Assistant, now live in the Adobe Firefly app, the all-in-one creative AI studio. Powered by Adobe's creative agent, Firefly AI Assistant lets you start with your vision,
Starting point is 00:20:50 just describe what you want, and shape the outcome as it takes form with the Assistant. The Assistant orchestrates multi-step workflows, drawing on 60-plus pro-grade tools across Adobe Creative Cloud apps, including Photoshop, Illustrator Premiere, Lightroom Express, and more to help bring your ideas to life. can also get started with creative skills, a growing library of pre-built workflows for common creative tasks like batch editing photos, creating mood boards, portrait retouching, and creating social variations. Every step the assistant takes is visible so you can refine, redirect, or take over at any time. You stay in the driver's seat as the creative director. Adobe Firefly AI assistant now in public beta.
Starting point is 00:21:35 See it today at firefly.adobie.com. Let's keep it going. Number two, auto connectors to Gmail and calendar. I'm using this all the time. I was actually having a conversation with someone yesterday on this exact thing, like, hey, why can AI, you know, just get all my responses yet? For one single inbox, it can't, right? So I hope in the future this rolls out to multiple inboxes,
Starting point is 00:22:09 but this is a pretty big change that's kind of getting slept on. Here's why. These connectors inside chat, GPT, are new for 2025. The problem is a lot of their access modes have changed under the radar and not really announced. Okay. So as an example, you can connect your Gmail and your Google calendar. And you have been able to now for multiple months. I believe it was earlier this summer that the Gmail and Google Calendar connectors for chat
Starting point is 00:22:48 GPT were released. So then it's pretty nice. If you give it access to, right, it can access all of that information. The problem was previously it was only available in deep research. So for me, I didn't want to have to wait three to 12 minutes to get answers about my email or my calendar. If any of you out there have emailed me before, you know I'm absolutely terrible responding to emails. It's because I get spammed on all my email accounts, you know, people wanting me to, you know,
Starting point is 00:23:18 get their crappy startup on the, on the show. Sorry. A lot of you have great startups, but there are so many people, you know, every day, oh, I have something better than literally. People are like, I have something better than chat GPT. You need me on the show and then they badger me. And I'm like, no. Anyways, I'm terrible at email.
Starting point is 00:23:37 there's so many things on there. So I've been able to kind of refine a prompt that helps, you know, giving it examples of what's good, like what's an email I care about, what's an email I don't care about and why. And then just saying like, hey, which ones have I not responded to, et cetera? But you can even just say something as simple as, you know, which emails from my Gmail inbox need my immediate response. All right. And it'll go through tell you and it'll even draft some example replies.
Starting point is 00:24:03 So like I said, the updates here. now it's instant. It is auto. You don't even have to click it. It is once you connect it, it is on by default. So that's great. So it used to be deep research only. And if I'm being honest, this is one of the few remaining reasons I was using Claude. So Anthropics Claude had their version of this as well. But you didn't have to use deep research for Gmail and Google calendar. So I was using that a lot. When it came out, I believe it It was April, May. It was one of the few reasons I was using Claude on an ongoing basis to just go through my email
Starting point is 00:24:43 and find things, right? I have all these, you know, email threads with like, you know, 80 different, you know, replies and 20 different people on the emails going in and out. And it's confusing for me. So I'd always just jump into Claude and be like, hey, catch me up on this. I forgot. What do I need to do? What are they asking?
Starting point is 00:25:00 Right. So now I can do that inside chat, GPT. So I hope they add a multi-inbox support soon. But I think this is one of if you struggle with email like me, this is one of the best ways I think to just be able to say, hey, where should I be spending my attention? If you get spammed all the time like me or just get a lot of, you know, legitimate business opportunities,
Starting point is 00:25:24 people just reaching out. If you're struggling, getting through your Gmail, it's a great way to do it. All right. Last but not least. Number one, custom instruction. for projects.
Starting point is 00:25:36 Let's go. This one I've literally been waiting for for a very long time. So I kind of explained a little bit earlier what custom instructions work. So it's similar to a system prompt. So let's say as an example, that you always want chat GPT to respond in a certain way. Maybe that's the brand voice of your company. Maybe you want it to reply in, you know, short faction. bullet points. Maybe you want it to be, you know, humorous and engaging. So unfortunately,
Starting point is 00:26:10 I think so many people use chat GPT for multiple different reasons, right? I use it for multiple different businesses. I use it for clients. I use it for things in my personal life. So custom instructions, aside from, yo, don't lie, which is kind of my zero fabrication custom instructions, the only ones that I use. I can't get custom instructions. I can't get custom instructions. that can only apply to certain chats. That was one of the biggest problems with custom instructions is if, if you used chat GPT for only one purpose, then custom instructions were helpful, right?
Starting point is 00:26:47 So you didn't have to, each time you went into a new chat, you didn't have to explain how you wanted to respond, the do's and don'ts, right? You kind of set your own custom rules. But let's say I, for a certain client,
Starting point is 00:26:58 I always want short, factual things with citations, right? But then if I want to use it for everyday AI to help me, you know, improve an email or something like that, I don't want it to go through and be short and factual bullet points with citations, right? So one of the biggest problems with custom instructions was, well, they applied unilaterally to every single chat. You couldn't turn them off or on on a different chat. It's, well, you either use it for literally everything or literally nothing. So now, again, barely announced. I think there was one little update in OpenA.I.
Starting point is 00:27:33 change log on this, but I think it is probably the number one feature that I myself have been looking forward to for multiple years that just hasn't come. So this is huge. So if you don't know projects, think of projects as folders, but a little bit better. So a lot of people think projects are just ways to organize your different chats, right? So instead of them all just being, you know, huddled on the left side, you know, in your sidebar and scrolls, rolling through dozens or in my case, you know, sometimes I'll have 50 plus chats a day inside chat, JVT. Instead, you can sort them in folders. But aside from just having them in folders, there's also some different in additional capabilities of using projects, which are really helpful.
Starting point is 00:28:23 So number one would be you can upload files that are accessible to any new chat in that project, which is extremely helpful. And also now there's project. memory. So this was rolled out a few months ago. So if you're using deep research, it can research across those projects. But now you have custom instructions. It's a little tricky though. So you have to kind of do this the right way. So don't just go in and create a project and start chatting because once you start that, you cannot change the memory. So you can only do this and do project only memory once you create a new project. So look for the gear icon. So it's kind of hidden, right? So when you create a new project, it's not going to be an option you even see. So before you
Starting point is 00:29:12 literally even create the project, go find the gear icon and then you have to set the memory. So by default, you can set it default. And by default, that's obviously the default trigger. And it says project can access memories from outside chats and vice versa or project only. So it says the project can acts can only access its own memories. Its memories are hidden from outside chats. That's the big thing. So now let's say, you know, you have, you know, you use chat GPT for two different businesses and for personal use.
Starting point is 00:29:43 So at bare minimum, just create three new folders, right? You know, company one, company two, personal. And it does, yes, it takes an extra five seconds when you go and start a new chat to go and click the project first. but it's going to end up saving you so much time. All right. So a couple more details and recap on this. And I think this is finally what makes custom instructions actually useful.
Starting point is 00:30:11 And even I said this, I've personally advised against custom instructions since the day they came out. Because I would say 90% in my interaction with people and I've trained like 13,000 people live on chat, GPD, I would say less than 10% use chat GPD for only one reason. So custom instructions for the most part, aside from if you're, you know, trying to use it to cut down on hallucinations, but if you're trying to get it to customize or personalize chat gbt, it hasn't been useful for hardly anyone. But custom instructions previously were applied unilaterally.
Starting point is 00:30:50 And now you have so much flexibility. And a lot of people are like, well, isn't this just what you would use GPs for? Yes. But with GPTs, it's, you know, I'll probably do a future episode going over the intricacies and the fine detailed differences between GPTs and folders. But one of the biggest ones is, well, GPD doesn't offer, GPDs don't offer a way to easily organize those chats, right? It's like, hey, I use this GPT for this a week ago. You can't go into your GPT and see all the different chats that have been used there. So there's just much more projects are much more flexible.
Starting point is 00:31:27 than GVTs with the organizational structure and also that deep research capability within a individual project, which is huge. All right. Now we've been through this. All right. Mark, I think Mark corrected. I think Mark said zero now. Someone here on LinkedIn saying this is such good information. Thank you.
Starting point is 00:31:52 You're welcome. All right. So let's end this show. Ah, just kidding. Like I said, bonus. There was two. After I already posted the preview for the show and I had my five, you know, overlooked chat GPT features done.
Starting point is 00:32:06 I was like, oh my gosh, I completely forgot about these two use cases for chat GBT's agent mode that were not out or available when chat GPT agent mode released. So I did two different shows on chat GPT agent mode. So if you're thinking like, Jordan, you already told me everything. No, I didn't. These are brand new updates. So if you want access to those, I have a little interactive website built with Gemini Canvas,
Starting point is 00:32:33 ready to go. There's AI features enabled, different use cases, examples, everything's in there. So if you liked these five and you're like, okay, well,
Starting point is 00:32:41 I want a cool way to look at these or use these, but you want these other two, make sure to repost this on LinkedIn. All right. So I'm going to go through probably in a couple of days and send this to everyone. So if you do want access, just go ahead and repost this on LinkedIn. and I will send you this.
Starting point is 00:32:57 So again, quick recap those five brand new updates that I think are getting slept on, flashcards with GPT quiz, custom personalities, advanced voice mode updates and access inside GPTs, auto connection for Gmail and Google Calendar in normal chat, chitvety chat, and then the projects getting access to custom instructions. I hope this was helpful. If so, please share this with your. your audience. If you're listening on the podcast, I appreciate it. As always, please follow, like and subscribe to the show. If you could leave us a rating on the podcast, I'd really appreciate
Starting point is 00:33:36 that. Takes 10 seconds. And then when you're done with that, please go to your everyday AI.com. Sign up for the free daily newsletter. We're going to be recapping everything from today's show. So if you miss anything, don't worry. It's got to be in there as well as everything you need to be the smartest person in AI at your company. Thanks for tuning in. Hope to see you back tomorrow and every day for more Everyday AI. you'll meet Firefly AI assistant now live in adobe Firefly the allman one creative AI studio just describe what you want to create in your own words and the assistant handles the rest orchestrating multi-step workflows across adobe creative cloud apps including Photoshop
Starting point is 00:34:17 premiere express and more in one conversational interface you direct the outcome while the assistant accelerates execution stand control with the ability to step in and refine at any time see it today at firefly.adobie.com. And that's a wrap for today's edition of Everyday AI. Thanks for joining us. If you enjoyed this episode, please subscribe and leave us a rating. It helps keep us going. For a little more AI magic, visit Your EverydayAI.com and sign up to our daily newsletter
Starting point is 00:34:53 so you don't get left behind. Go break some barriers and we'll see you next time.

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.