WEBVTT 00:00:00.000 --> 00:00:02.500 You know, many professionals, maybe even you 00:00:02.500 --> 00:00:04.780 listening right now, are paying quite a bit each 00:00:04.780 --> 00:00:08.160 year, hundreds maybe, for various AI tools. Yeah, 00:00:08.220 --> 00:00:10.580 subscriptions for research tools, AI writers, 00:00:10.859 --> 00:00:13.740 image generators. It adds up fast. It's become 00:00:13.740 --> 00:00:16.879 this sort of background cost. But when we looked 00:00:16.879 --> 00:00:19.179 into the sources for today, there's this big 00:00:19.179 --> 00:00:22.399 kind of surprising twist. Right. Google's actually 00:00:22.399 --> 00:00:25.179 rolled out this whole integrated AI ecosystem. 00:00:25.699 --> 00:00:29.429 And honestly, a lot of it is... arguably better 00:00:29.429 --> 00:00:31.429 than the paid stuff. And it's free. Completely 00:00:31.429 --> 00:00:34.090 free. It's almost an unfair advantage waiting 00:00:34.090 --> 00:00:36.229 to be used. Exactly. We're talking about an AI 00:00:36.229 --> 00:00:40.170 super team here. Think Gemini, AI Studio, Notebook 00:00:40.170 --> 00:00:42.950 LM, Opal, and this thing called AI Mode. We've 00:00:42.950 --> 00:00:44.909 got the playbook. So our mission today is pretty 00:00:44.909 --> 00:00:48.490 clear. Let's unpack maybe 30 or more real world 00:00:48.490 --> 00:00:50.549 professional ways you can use these free tools. 00:00:50.689 --> 00:00:53.090 We're talking automation, content creation, saving 00:00:53.090 --> 00:00:55.750 serious time. Okay. Let's kick things off with 00:00:55.750 --> 00:00:58.539 the big one. Gemini. The sources call it the 00:00:58.539 --> 00:01:00.899 Swiss army knife of AI, which feels about right. 00:01:01.060 --> 00:01:03.859 It's your main conversational AI. But the killer 00:01:03.859 --> 00:01:05.620 feature, the thing that really sets it apart, 00:01:05.780 --> 00:01:09.180 is how deeply it plugs into Google Workspace. 00:01:09.659 --> 00:01:13.379 Your drive, your email, even YouTube. That context 00:01:13.379 --> 00:01:15.620 makes all the difference. Okay. Okay, use case 00:01:15.620 --> 00:01:18.980 one. Your personal email assistant, you just 00:01:18.980 --> 00:01:21.799 flip on the Workspace integration, and you can 00:01:21.799 --> 00:01:24.579 literally ask it, summarize my important emails 00:01:24.579 --> 00:01:27.299 from the last two days. Boom. Prioritize list. 00:01:27.719 --> 00:01:30.079 And it'll draft replies that sound like you. 00:01:30.180 --> 00:01:32.879 Because it knows you from your other emails and 00:01:32.879 --> 00:01:35.579 docs. It's not just generic stuff. Exactly. The 00:01:35.579 --> 00:01:38.000 personal context is key. Then there's the YouTube 00:01:38.000 --> 00:01:40.549 intelligence analyst. This is pretty cool. Yeah, 00:01:40.569 --> 00:01:42.790 this one's wild. You take, say, a two -hour keynote 00:01:42.790 --> 00:01:45.010 video, something you just don't have time for. 00:01:45.189 --> 00:01:47.450 Right. You feed Gemini the link and say, pull 00:01:47.450 --> 00:01:49.769 out the key insights relevant to my Project X 00:01:49.769 --> 00:01:52.329 goals. It doesn't just summarize. It connects 00:01:52.329 --> 00:01:54.349 the dots for you. That's analysis, not just a 00:01:54.349 --> 00:01:56.870 transcript. The time -saving is huge. Oh, yeah. 00:01:56.950 --> 00:02:00.569 And how about this? The screen recording to documentation 00:02:00.569 --> 00:02:03.870 machine. Get this. You record your screen doing 00:02:03.870 --> 00:02:06.209 some complex tasks, no audio needed, just the 00:02:06.209 --> 00:02:08.289 video. You upload that silent clip, maybe five 00:02:08.289 --> 00:02:11.169 minutes long. And Gemini spits out a detailed 00:02:11.169 --> 00:02:14.669 step -by -step written guide, an SOP, just like 00:02:14.669 --> 00:02:17.289 that. Think of the hours saved on technical writing. 00:02:17.629 --> 00:02:20.949 It turns visual how -to into actual instructions. 00:02:21.169 --> 00:02:24.930 And for sharing complex stuff quickly. The research 00:02:24.930 --> 00:02:27.949 to visual storytelling pipeline. Got a dense 00:02:27.949 --> 00:02:31.069 report. Use the Canvas feature. One click, basically. 00:02:31.250 --> 00:02:34.289 And it helps turn that text into a clean, professional 00:02:34.289 --> 00:02:36.810 infographic. Makes complex data easy to grasp 00:02:36.810 --> 00:02:38.840 for... you know anyone you need to show it to 00:02:38.840 --> 00:02:41.719 so that deep workspace integration yeah what's 00:02:41.719 --> 00:02:43.460 the fundamental change it brings compared to 00:02:43.460 --> 00:02:46.400 other ais well it's that personal context right 00:02:46.400 --> 00:02:48.800 from your stuff emails docs other tools just 00:02:48.800 --> 00:02:50.879 don't have that okay so gemini's the all -rounder 00:02:50.879 --> 00:02:54.629 yeah next up ai studio This feels more like Google's 00:02:54.629 --> 00:02:57.129 hidden R &D lab. Yeah, the hidden playground 00:02:57.129 --> 00:02:58.909 is a good name for it. It's where you get early 00:02:58.909 --> 00:03:01.169 free access to some of their most powerful cutting 00:03:01.169 --> 00:03:04.530 edge models. Stuff like imaging for images, advanced 00:03:04.530 --> 00:03:06.930 text speech. And it's not just playing around. 00:03:07.150 --> 00:03:09.590 It's like a full on production studio. The quality 00:03:09.590 --> 00:03:11.750 you can get is kind of nuts. Let's talk about 00:03:11.750 --> 00:03:13.819 the professional voice actor use case. Right. 00:03:13.919 --> 00:03:15.939 The text -to -speech. It's seriously good. Like, 00:03:15.960 --> 00:03:18.219 broadcast quality good. You feed it a script. 00:03:18.439 --> 00:03:20.199 And it generates this really natural sounding 00:03:20.199 --> 00:03:23.139 audio, perfect for training videos, internal 00:03:23.139 --> 00:03:25.840 announcements. You skip the whole cost and hassle 00:03:25.840 --> 00:03:28.199 of hiring voice actors. You can scale it. Get 00:03:28.199 --> 00:03:31.360 this. The multi -speaker podcast generator. Wait, 00:03:31.360 --> 00:03:34.110 what? Yeah. You take something static, like an 00:03:34.110 --> 00:03:37.490 internal FAQ document. Boring, right? Totally. 00:03:37.810 --> 00:03:40.189 AI Studio can turn that into a conversation. 00:03:40.250 --> 00:03:43.569 You assign roles, like HR manager and new employee, 00:03:43.689 --> 00:03:47.129 and it generates a Q &A dialogue with two distinct 00:03:47.129 --> 00:03:50.150 natural voices. Seriously, that capability is 00:03:50.150 --> 00:03:53.770 just there and free. That's a game changer for 00:03:53.770 --> 00:03:56.189 making internal POMs or training actually engaging. 00:03:56.449 --> 00:03:59.099 Whoa. I mean, imagine scaling that voice actor 00:03:59.099 --> 00:04:01.740 thing across like a billion training questions 00:04:01.740 --> 00:04:05.099 inside a huge company. That's just massive efficiency. 00:04:05.120 --> 00:04:06.840 And it's not just audio, right? The visuals. 00:04:07.139 --> 00:04:09.139 Equally impressive. The creative asset generator 00:04:09.139 --> 00:04:11.639 uses their image and model for really high -res 00:04:11.639 --> 00:04:14.460 images. The pro tip from the sources. Find an 00:04:14.460 --> 00:04:16.740 image you like, analyze a screenshot of it to 00:04:16.740 --> 00:04:19.300 get a detailed prompt. Ah, reverse engineer the 00:04:19.300 --> 00:04:22.500 prompt. Yeah. Clever. Yeah. Then use that detailed 00:04:22.500 --> 00:04:25.120 prompt in AI Studio to create something new in 00:04:25.120 --> 00:04:28.079 a similar style. And for product folks, there's 00:04:28.079 --> 00:04:30.740 the live UX feedback session. How does that work? 00:04:30.939 --> 00:04:33.180 You share your screen showing your app or website 00:04:33.180 --> 00:04:36.660 with the AI. It acts like a potential user giving 00:04:36.660 --> 00:04:40.439 you brutally honest feedback in real time. Hmm. 00:04:40.600 --> 00:04:43.519 I hesitated here. This button wasn't clear. Finds 00:04:43.519 --> 00:04:46.339 friction points instantly. Okay, but if these 00:04:46.339 --> 00:04:48.600 models are sometimes experimental, stuff straight 00:04:48.600 --> 00:04:51.519 from the lab, is the output actually reliable 00:04:51.519 --> 00:04:54.420 enough for, you know, real business use? Yeah, 00:04:54.459 --> 00:04:56.319 that's the surprising part. The output quality 00:04:56.319 --> 00:04:58.379 is professional grade. Definitely good enough 00:04:58.379 --> 00:05:00.139 for marketing materials, corporate training, 00:05:00.259 --> 00:05:02.319 that kind of thing. Okay, let's shift gears to 00:05:02.319 --> 00:05:05.540 research. Notebook LM. The sources call this 00:05:05.540 --> 00:05:08.079 the research powerhouse. And the key thing here, 00:05:08.139 --> 00:05:10.379 the absolute most important concept, is source 00:05:10.379 --> 00:05:12.579 grounding. Explain that. What does source grounding 00:05:12.579 --> 00:05:15.459 mean? It means Notebook LM works only with the 00:05:15.459 --> 00:05:18.220 documents you upload. Up to 50 documents on the 00:05:18.220 --> 00:05:20.680 free plan. It doesn't go searching the web. It 00:05:20.680 --> 00:05:22.420 doesn't pull from its general training data. 00:05:22.680 --> 00:05:25.379 Its knowledge is limited to your sources. Ah, 00:05:25.560 --> 00:05:28.300 okay. And why is that such a big deal? Because 00:05:28.300 --> 00:05:31.759 it drastically cuts down the risk of the AI just 00:05:31.759 --> 00:05:34.199 making stuff up, those hallucinations everyone 00:05:34.199 --> 00:05:36.819 worries about. If you need reliable, accurate 00:05:36.819 --> 00:05:39.720 synthesis for professional work, using only your 00:05:39.720 --> 00:05:42.579 validated docs is crucial. It keeps things factual. 00:05:42.879 --> 00:05:46.500 Got it. So accuracy and reliability are paramount 00:05:46.500 --> 00:05:48.199 here. Exactly, which lets you do some pretty 00:05:48.199 --> 00:05:51.519 powerful things like... The corporate espionage 00:05:51.519 --> 00:05:54.279 agent use case. Sounds intriguing. You basically 00:05:54.279 --> 00:05:57.279 upload maybe 50 URLs from competitor websites 00:05:57.279 --> 00:05:59.579 or their white papers, annual reports, whatever 00:05:59.579 --> 00:06:02.980 you can find. Notebook LM reads them all and 00:06:02.980 --> 00:06:04.939 generates these really comprehensive analyses. 00:06:05.180 --> 00:06:07.259 What are their strategies? Who are they targeting? 00:06:07.439 --> 00:06:09.060 Are there gaps in the market they're missing? 00:06:09.279 --> 00:06:12.050 Wow. That's automated competitive intel that 00:06:12.050 --> 00:06:14.689 would normally take an analyst ages. Weeks. Yeah. 00:06:14.750 --> 00:06:17.889 And for internal stuff, picture the project podcast 00:06:17.889 --> 00:06:20.189 generator. Okay. You feed it all the documents 00:06:20.189 --> 00:06:23.290 for a big project meeting notes, specs, emails, 00:06:23.529 --> 00:06:26.029 the works. It then creates an audio discussion 00:06:26.029 --> 00:06:28.430 about the project. Like a simulated meeting? 00:06:28.470 --> 00:06:30.470 Kind of. And you can tell it who should be talking. 00:06:30.649 --> 00:06:33.050 Like, have a skeptical stakeholder debate an 00:06:33.050 --> 00:06:35.759 optimistic project manager. It helps you anticipate 00:06:35.759 --> 00:06:37.879 tricky conversations. That's actually really 00:06:37.879 --> 00:06:40.160 useful for prepping for real meetings. Definitely. 00:06:40.259 --> 00:06:42.620 And maybe the most practical tool for any team. 00:06:43.120 --> 00:06:45.600 The meeting secretary. Oh, I think I know where 00:06:45.600 --> 00:06:47.879 this is going. Yep. Upload the transcripts from 00:06:47.879 --> 00:06:50.779 your recent meetings. The AI automatically identifies 00:06:50.779 --> 00:06:53.199 all the action items, who's responsible, and 00:06:53.199 --> 00:06:55.660 the deadlines, and puts it all into an e -table. 00:06:55.800 --> 00:06:58.220 Okay, that alone could save hours every week. 00:06:58.439 --> 00:07:01.600 Simple, but powerful. Now, quick question on 00:07:01.600 --> 00:07:04.300 Notebook LM. Since it's working with potentially 00:07:04.300 --> 00:07:06.980 sensitive internal documents, even if it's source 00:07:06.980 --> 00:07:10.240 grounded, is Google using that data to train 00:07:10.240 --> 00:07:12.899 its bigger models? Good question. The sources 00:07:12.899 --> 00:07:15.560 are really clear on this. No, Notebook LM is 00:07:15.560 --> 00:07:18.379 treated as a private workspace. Your uploaded 00:07:18.379 --> 00:07:20.800 documents are not used to train Google's foundation 00:07:20.800 --> 00:07:23.279 models. OK, that's reassuring. So last segment, 00:07:23.379 --> 00:07:25.180 we're looking at the future of actually doing 00:07:25.180 --> 00:07:28.920 things with AI, Opal and AI mode. Right, so Opal 00:07:28.920 --> 00:07:31.000 first. It's currently in beta, US only for now, 00:07:31.139 --> 00:07:33.920 but it's basically a no -code AI builder. Think 00:07:33.920 --> 00:07:36.120 drag and drop for creating automated AI workflows. 00:07:36.399 --> 00:07:38.220 So you don't need to be a programmer to chain 00:07:38.220 --> 00:07:40.939 AI tasks together. Exactly. Opal is like the 00:07:40.939 --> 00:07:43.300 factory floor where you build custom AI assembly 00:07:43.300 --> 00:07:46.199 lines. You can pull in context from Gemini, use 00:07:46.199 --> 00:07:48.980 Imogen from AI Studio for visuals, all linked 00:07:48.980 --> 00:07:50.839 together in a repeatable process. Okay, give 00:07:50.839 --> 00:07:52.819 me an example. The automated site audit report. 00:07:53.040 --> 00:07:56.000 You build a workflow in Opal. Step one, input 00:07:56.000 --> 00:08:00.459 a website URL. Step two, Opal triggers various 00:08:00.459 --> 00:08:04.439 AI checks, SEO, maybe content analysis. Step 00:08:04.439 --> 00:08:07.120 three, it generates a professional -looking Google 00:08:07.120 --> 00:08:10.199 Doc report with recommendations. Automatically, 00:08:10.199 --> 00:08:12.639 just from the URL. Yep, zero code needed. Or 00:08:12.639 --> 00:08:15.500 the automated marketing asset generator. Faster 00:08:15.500 --> 00:08:18.019 marketing content. Much faster. Upload a product 00:08:18.019 --> 00:08:20.040 photo, tell it about your target audience. Opal 00:08:20.040 --> 00:08:22.160 runs a workflow that generates images, headlines, 00:08:22.420 --> 00:08:24.480 social media copy, maybe even ad variations. 00:08:24.699 --> 00:08:27.120 Full set of assets ready to go in like under 00:08:27.120 --> 00:08:29.959 a minute. Okay, that's impressive. I have to 00:08:29.959 --> 00:08:32.360 admit, I still wrestle sometimes with getting 00:08:32.360 --> 00:08:34.799 complex prompts right, you know, prompt drift 00:08:34.799 --> 00:08:36.879 and making different AI steps talk to each other. 00:08:37.000 --> 00:08:39.320 Yeah, building those chains manually can be tricky. 00:08:39.580 --> 00:08:42.200 Opal simplifies it massively with that visual 00:08:42.200 --> 00:08:45.139 drag and drop interface, makes complex automation 00:08:45.139 --> 00:08:48.019 accessible. So Opal handles the custom building. 00:08:48.240 --> 00:08:51.940 What about AI mode? AI mode is about supercharging 00:08:51.940 --> 00:08:54.379 Google search itself, making it less about just 00:08:54.379 --> 00:08:56.799 blue links and more about... direct answers and 00:08:56.799 --> 00:08:59.179 analysis. Like the instant comparison table example. 00:08:59.460 --> 00:09:01.840 Exactly. Instead of opening 10 browser tabs to 00:09:01.840 --> 00:09:03.980 compare, say, three different software products, 00:09:04.159 --> 00:09:07.320 you just ask search in AI mode. It pulls the 00:09:07.320 --> 00:09:10.120 info features, pricing, common complaints, and 00:09:10.120 --> 00:09:12.200 gives you an organized table right there on the 00:09:12.200 --> 00:09:14.580 results page. That saves a ton of clicking and 00:09:14.580 --> 00:09:17.059 mental energy. For sure. And there's also precision 00:09:17.059 --> 00:09:19.970 source search. This is cool. You can tell AI 00:09:19.970 --> 00:09:22.509 mode to search only within a specific website, 00:09:22.730 --> 00:09:25.269 like a company's blog or documentation pages. 00:09:25.529 --> 00:09:26.970 Ah, so you're not wading through the whole internet. 00:09:27.399 --> 00:09:29.519 Right. You can ask it something like, summarize 00:09:29.519 --> 00:09:31.960 the main themes on the DigitalOcean blog about 00:09:31.960 --> 00:09:34.720 Kubernetes over the last year, and it'll do that, 00:09:34.779 --> 00:09:36.860 pulling info just from that site and giving you 00:09:36.860 --> 00:09:39.100 categorized links to verify. Okay, so you have 00:09:39.100 --> 00:09:42.279 Opal for building custom workflows and AI mode 00:09:42.279 --> 00:09:44.539 for getting smarter, faster, more analyzed search 00:09:44.539 --> 00:09:47.200 results. What's the big strategic payoff when 00:09:47.200 --> 00:09:49.139 you put those two together? Well, you can build 00:09:49.139 --> 00:09:51.769 these really specific. automated workflows in 00:09:51.769 --> 00:09:54.590 opal and they can be constantly fed and updated 00:09:54.590 --> 00:09:58.210 with real -time fact -check data pulled intelligently 00:09:58.210 --> 00:10:01.490 via ai mode search custom automation granted 00:10:01.490 --> 00:10:03.909 in current reality right makes sense so let's 00:10:03.909 --> 00:10:06.549 pull it all together this unfair advantage the 00:10:06.549 --> 00:10:08.669 sources really point to four key things that 00:10:08.669 --> 00:10:11.830 make this google ai ecosystem so powerful especially 00:10:11.830 --> 00:10:14.340 because it's free First, the ecosystem effect. 00:10:14.580 --> 00:10:16.519 Yeah, how everything just fits together. Gemini 00:10:16.519 --> 00:10:19.379 knows your stuff. AI Studio makes things. Notebook 00:10:19.379 --> 00:10:21.919 LM checks fax against your docs. Opal automates 00:10:21.919 --> 00:10:24.419 it all. It's integrated, like Lego blocks of 00:10:24.419 --> 00:10:28.100 data and capability, kind of. Second, the freedom 00:10:28.100 --> 00:10:31.019 factor. The free tiers are generous. You don't 00:10:31.019 --> 00:10:32.840 have that usage anxiety worrying about running 00:10:32.840 --> 00:10:35.419 up a bill. You can just experiment, try things, 00:10:35.539 --> 00:10:38.019 build stuff without that pressure. That encourages 00:10:38.019 --> 00:10:40.539 innovation. Definitely. Third was the cutting 00:10:40.539 --> 00:10:42.840 edge advantage. You're often getting access to 00:10:42.840 --> 00:10:44.759 Google's newest models, sometimes before they're 00:10:44.759 --> 00:10:46.799 widely released, like that nano banana image 00:10:46.799 --> 00:10:48.860 model mentioned. You stay ahead of the curve. 00:10:49.039 --> 00:10:52.919 Yep. And finally, number four, production ready 00:10:52.919 --> 00:10:55.919 results. This isn't just theoretical or demo 00:10:55.919 --> 00:10:59.500 wear. The audio from AI Studio, broadcast quality, 00:10:59.740 --> 00:11:02.419 the images, high res, the reports from Notebook 00:11:02.419 --> 00:11:05.220 LM, structured and reliable. It's stuff you can 00:11:05.220 --> 00:11:07.340 actually use in your business today. So the flip 00:11:07.340 --> 00:11:10.440 side, the hidden cost of not tapping into this, 00:11:10.580 --> 00:11:12.679 it's pretty straightforward, isn't it? Yeah. 00:11:12.720 --> 00:11:14.840 You risk falling behind competitors who are using 00:11:14.840 --> 00:11:17.019 it. You're likely wasting money on paid tools 00:11:17.019 --> 00:11:19.659 that this free suite can replace. And you're 00:11:19.659 --> 00:11:21.799 definitely spending time on manual tasks that 00:11:21.799 --> 00:11:24.159 could and probably should be automated by now. 00:11:24.299 --> 00:11:26.399 It feels like the conversation is shifting. It's 00:11:26.399 --> 00:11:29.620 not just, is free AI good enough anymore? With 00:11:29.620 --> 00:11:32.539 this level of integration, context, and quality. 00:11:33.580 --> 00:11:36.139 The free ecosystem is often winning, especially 00:11:36.139 --> 00:11:38.759 for certain tasks. Absolutely. The context from 00:11:38.759 --> 00:11:41.120 workspace and the source grounding in Notebook 00:11:41.120 --> 00:11:44.159 LM, plus the production quality, it's a powerful 00:11:44.159 --> 00:11:46.470 combination that's hard to beat. paid or not. 00:11:46.610 --> 00:11:49.889 So Google's basically put this incredibly powerful 00:11:49.889 --> 00:11:53.009 AI toolkit on the table, free of charge. The 00:11:53.009 --> 00:11:55.730 keys to the AI kingdom, almost, without the massive 00:11:55.730 --> 00:11:58.269 budget requirement. Which leaves just one real 00:11:58.269 --> 00:11:59.889 question for you, the listener. Free, you can 00:11:59.889 --> 00:12:02.669 build with it. Exactly. What specific part of 00:12:02.669 --> 00:12:04.789 your work, your research, your content creation, 00:12:05.009 --> 00:12:07.669 what will you transform or automate first with 00:12:07.669 --> 00:12:08.370 this free arsenal?