Speaker 0: the era of the passive chatbot is dead, and google just turned your office suite into a fleet of autonomous agents. everybody's getting into the autonomous agent world, john. we aren't just talking about ai writing your emails anymore. we're looking at the system that can audit invoices, build its own apps, and fill spreadsheets nine times faster than a human. google's workspace becomes agentic ai. so google has launched workspace intelligence, a unified layer that connects gmail, drive, and sheets to perform autonomous tasks. it no longer just suggesting text. it's executing multistep workflows like a digital employee. this effectively turns every employee into a manager of ai agents, potentially eliminating the boring 60% of office admin work overnight. what do you think about this, john? Speaker 1: yeah. so not surprising. Speaker 0: think they're just an odd, also, ron, trying to catch up? Speaker 1: i think that this is a smart move by google, because everyone in their, everyone uses or most people use gmail and docs. so, integrating something that's more useful than just writing emails or summarizing documents makes sense. and gemini is pretty decent, as far as being a multimodal ai kind of family. right? so, i like it because it just integrates seamlessly if that makes sense. right? just, it's you don't have to add on anything. you don't have to install anything. at least, i don't think so. you have to enable some things, but, this makes sense now that they have a working, a good solid llm, if that makes sense. Speaker 0: yeah. so you you use more Speaker 1: of these automation tools than i do. so what do you think? Speaker 0: well, i've just been using more and more google workspace automation over the last, few days just because i have all these other automation tools that i use, and i can't, you know, test and run every one of them that come out. but i like to stay on top of it, and i do use google workspace. and, yeah, there really seems to be two dominant office platforms, you know, the, google workspace and then microsoft. and, you know, the startup world seems to be dominated by google, whereas the enterprise seems to be dominated by microsoft. now, i mean, people can argue with me here and there, but that seems to be the split. then, you know, students and whatnot, you'll mostly use, google because they can get it for free or low end or what have you. but, anyway, think of this like an update from a gps that gives you directions and moving to a self driving car. and all the ai told you how to write the email or told you how to organize the data, where identik ai actually puts its hands on the wheel, drives to the destination, and handles the paperwork once it gets there. yeah. imagine this. yeah. go ahead. Speaker 1: no. no. Speaker 0: no. you can immediately now paste messy brain dump of text that you've, dictated or grabbed from a bunch of emails or whatever into sheets and have it instantly converted into professional formatted, table as an example. Speaker 1: yeah. so the the the breakthrough here that i'm seeing here is that it's the connection between all of these, like you said. right? so it's a knowledge graph of all your work where ai understands all these relationships. so this is a perfect, like, think of it like the one stop shop for everything that you're working on the only problem is that it's going to be, it's going to be disconnected between gmail accounts, i think. right? Speaker 0: well, i don't know what the enterprise version versus teams versus individual are gonna look like, and what will be shared or what won't be. and so always a little funky when you have a corporate account with, google and some of these things because they're very pernickety about who gets access to data and stuff. so i don't know whether i would assume at some time, admin can have global templates, global knowledge base that they can point to, but i have not played with it yet. so i don't wanna make those assumptions. yeah. yeah. fair enough. specs are nine times faster data entry for a 100 cell tasks like you with, using phil with gemini saying, oh, just go put this in. five times faster migration speed from microsoft three sixty five to google workspace. they've got ai to help you migrate faster, i guess. and, they have an 8.5% year over year growth in ai generated notes using, usage within google meet. and i think that's gonna get more and more and more. i noticed lots of other tools now have, just recently put a updated full fledged google meet agent. so, for example, with openclaw, my openclaw can attend meetings for me now or with me, take notes, do action items, take actions right there, live in the session based on what's going on. like, it's crazy what things can do. so the more you start integrating these features, because that's a third party product doing it. so if you can talk to your spreadsheet and say, you know what? bob sent me some numbers last week, and jane did the week before. go get those status reports and put them in the spreadsheet here and summarize them for me, and it goes checks your email, grabs the stuff, and does it for you. you just start realizing that that's something you can tell it to do. great. but the active ai agents will eventually say, you know what? from now on, every time this type of information comes in, i want you to build a report from this or add it to this report, and that stuff just starts happening. and so those spreadsheets and workflows and monthly reports start getting built, and things start happening automatically that you don't understand. and let me give you an example of it. i installed an open client in the cloud, filocloud, gave it access to my google. it didn't really do anything with it. but it automatically responded to you. it changed our calendar appointment. it responded to other people. it set a potential client a quote before i realized it was doing anything. i didn't it just started doing stuff. and, like, i had to rein it in a little bit because it made a couple of mistakes. it moved your meeting to a comp place where i had a conflict rather than checking for conflict. and so you gotta only because i didn't tell it these things. Speaker 1: yeah. Speaker 0: but i'm just giving you, people out there listening, an example of all of a sudden you have a digital avatar assistant that will be thinking like you as you train it along and will do these things for you automatically so your work becomes more productive, more, more. you'll be you won't have to do the grunt work, the entry, putting all the stuff together. Speaker 1: yeah. i don't know if you remember. we might have talked about this before, but, if there there was someone who had asked ai, the replacement on a per token basis for one person and the cost for that. and there was millions of tokens, right, for someone's job, and i think the cost was under $10 or under $5. it's ridiculous. Speaker 0: yeah. well, we are going there, but let i'm gonna correct you on that right now. so, did you hear the ceo of nvidia talking? you know, when he's interviewing developers at the end of the year and doing the reviews, he's gonna ask them what was their token spend. and if someone says $5,000, he's gonna fire them immediately. he said your not token spend better be close to $50,000 or higher or you're you shouldn't be working here. so we need to and understand that the employees are not gonna be $10 a month. yeah. the whole thing won't run. like, you have to start understand there's some realistic costs around that. and but the gains are better. and once you get it locked in in these things, the gains are better. so we'll have to see how that goes. but if it's built in, for example, it's built into these things. so this is why learn leaning more to some of these built in tools because they don't use your tokens usually. yeah. they use or they use what i'm saying. you're you're already budgeted tokens versus many of the api tools which you're spending per usage all the time, and that really adds up versus saying it's all part of my subscription. i get these agents to do a hundreds hours of work for me for next to nothing because it's all part of my great. but that will only last for so long, because we're seeing it all over the place. everybody's raising their prices because they're losing money on the compute and the power, and they can't serve the audience as well. but i'm, really excited about how everybody's getting into this, a agentic workflow business. you know? with back to google, this system used agentic workflows which rely on iterative loops. it critiques its own draft and corrects its logic layer before you ever see the first version. to lure enterprise giants, google introduced sovereign data locking, allowing companies to legally tether their ai training data strictly to eu or us soil to bypass strict, regulations. what a look. yeah. i guess i don't know because i didn't check that one, but i would imagine if you built that in, you might as well build it in for other bigger countries as well. Speaker 1: might as well. yeah. yeah. Speaker 0: the skills feature in workspace studio essentially brings our robotic process automation to the masses. tech that used to cost a 100 k in enterprises licenses now drag and drop features in a spreadsheet. so it's building agent drag and drop agent builders right in the platform, which is potentially gonna change the game. but, you know, the agentic model requires deep access to your private data to be effective. if google's ai control center has more even one security lapse, the ai's contextual understanding becomes a blueprint for a massive data breach. Speaker 1: yeah. Speaker 0: yeah. so what do you think, john? are you gonna be jumping in and trusting this? Speaker 1: i think i'm gonna give it a shot once once it becomes available to me. yeah. for sure. Speaker 0: so google isn't just fighting your subscription. they're fighting to be the operating system, for how your business thinks. so, chad, would you, like, trust an ai skill to automatically pay your company's invoices that reads your emails and your google drives and responds and does things for you, or is that a step too far? let us know in the comments below.