WEBVTT 00:00:00.000 --> 00:00:02.500 Okay, so the AI revolution, it's given creators, 00:00:02.600 --> 00:00:06.139 well, incredible power, beat. But getting access 00:00:06.139 --> 00:00:09.019 to that power often means paying up a lot. We're 00:00:09.019 --> 00:00:12.880 seeing folks spend $100, maybe $200 every month 00:00:12.880 --> 00:00:15.300 just for the basic tools. You know, MidJourney, 00:00:15.380 --> 00:00:18.980 11 Labs, ChatGPT +, the essentials. But then 00:00:18.980 --> 00:00:21.440 Google comes along. And they've released some 00:00:21.440 --> 00:00:24.960 seriously powerful AI, stuff that feels enterprise 00:00:24.960 --> 00:00:28.440 level. We're talking advanced image tools, custom 00:00:28.440 --> 00:00:31.660 AI agents, deep research capabilities. And here's 00:00:31.660 --> 00:00:34.520 the kicker. completely free so this deep dive 00:00:34.520 --> 00:00:37.020 it's about taking that free arsenal from google 00:00:37.020 --> 00:00:39.780 and making it your well your secret weapon your 00:00:39.780 --> 00:00:41.740 competitive advantage welcome to the deep dive 00:00:41.740 --> 00:00:43.780 yeah today we're cracking open google's let's 00:00:43.780 --> 00:00:46.119 call it cost effective strategy pretty interesting 00:00:46.119 --> 00:00:47.539 we'll look at the main free tools they're offering 00:00:47.539 --> 00:00:50.340 gemini 2 .5 image stuff notebook lm and these 00:00:50.340 --> 00:00:52.640 things called gems but the really crucial part 00:00:52.640 --> 00:00:54.340 we're going to give you six strategic ways to 00:00:54.340 --> 00:00:56.560 actually combine these tools make them work together 00:00:56.560 --> 00:00:58.460 this really is about competing on your ideas 00:00:58.460 --> 00:01:00.439 you know not just how much money you can throw 00:01:00.439 --> 00:01:02.500 subscriptions yeah Okay, let's really unpack 00:01:02.500 --> 00:01:04.799 that financial side first. It's a real hurdle, 00:01:04.900 --> 00:01:07.780 this cost crisis in AI, as we're calling it. 00:01:07.799 --> 00:01:10.099 For anyone starting out, a small business, a 00:01:10.099 --> 00:01:13.420 creator, those monthly bills, they stack up really, 00:01:13.500 --> 00:01:15.700 really fast, painfully fast. Oh, they absolutely 00:01:15.700 --> 00:01:17.659 do. You want good images, right? Professional 00:01:17.659 --> 00:01:19.379 looking stuff. That's usually going to run you, 00:01:19.439 --> 00:01:23.000 what, $10 to maybe $60 a month, depending on 00:01:23.000 --> 00:01:26.129 the platform. Then video tools. Essential for 00:01:26.129 --> 00:01:28.430 social media these days. They often start around 00:01:28.430 --> 00:01:30.510 $12, but boy, they climb quickly from there. 00:01:30.590 --> 00:01:33.109 And voice generation. If you need clean narration, 00:01:33.209 --> 00:01:35.870 maybe for training videos or audio articles, 00:01:36.090 --> 00:01:39.030 you're easily looking at $20 to $50 a month for 00:01:39.030 --> 00:01:41.390 the really high quality output. Then add another 00:01:41.390 --> 00:01:44.349 $20 for your premium chatbot, like a ChatGPT 00:01:44.349 --> 00:01:46.849 Plus or Pro version. It's just a huge barrier 00:01:46.849 --> 00:01:49.629 to entry, seriously. A basic, kind of modest 00:01:49.629 --> 00:01:53.510 AI toolkit. Easily $100, maybe even $200 a month. 00:01:53.629 --> 00:01:55.609 And that's money that... frankly, could be going 00:01:55.609 --> 00:01:57.750 towards other essentials like your website hosting 00:01:57.750 --> 00:02:01.109 or actual marketing spend. It's a massive weight 00:02:01.109 --> 00:02:03.549 for new creators right off the bat. And it's 00:02:03.549 --> 00:02:05.129 not just the money, is it? There's also this 00:02:05.129 --> 00:02:08.090 subscription fatigue. It's a real thing. Just 00:02:08.090 --> 00:02:10.389 the mental energy of tracking all those different 00:02:10.389 --> 00:02:13.069 accounts, managing billing, learning different 00:02:13.069 --> 00:02:16.110 interfaces. It's a distraction. So Google rolling 00:02:16.110 --> 00:02:19.210 out these free alternatives, it removes that 00:02:19.210 --> 00:02:21.870 decision paralysis. You can just. Try things. 00:02:22.389 --> 00:02:24.509 Experiment. Integrate them into your workflow. 00:02:24.610 --> 00:02:27.389 Build stuff without constantly worrying about 00:02:27.389 --> 00:02:30.509 the budget blowing up. Right. So putting aside 00:02:30.509 --> 00:02:32.810 the obvious cost savings for a second, what would 00:02:32.810 --> 00:02:34.629 you say is the biggest functional benefit of 00:02:34.629 --> 00:02:36.229 these free tools, especially for someone just 00:02:36.229 --> 00:02:38.650 starting out? I'd say they remove that decision 00:02:38.650 --> 00:02:41.729 paralysis. They let you experiment fast, try 00:02:41.729 --> 00:02:43.789 things for free, and figure out your workflow 00:02:43.789 --> 00:02:45.569 without the financial pressure. Yeah, that freedom 00:02:45.569 --> 00:02:47.949 to experiment is huge. Okay, let's dive into 00:02:47.949 --> 00:02:49.509 the visual side first. Google's offering some 00:02:49.509 --> 00:02:51.590 pretty professional -grade free image editing 00:02:51.590 --> 00:02:55.250 now. It's part of the Gemini 2 .5 Flash image 00:02:55.250 --> 00:02:58.389 model. You access it through the Google AI Studio. 00:02:58.590 --> 00:03:00.469 We'll just call it Gemini Image for simplicity. 00:03:00.729 --> 00:03:02.870 Though, yeah, internally, it apparently had a 00:03:02.870 --> 00:03:06.580 very curious codename for a while. Softly. A 00:03:06.580 --> 00:03:09.259 very curious code name. But the results aren't 00:03:09.259 --> 00:03:11.939 curious. They're actually quite polished. Let's 00:03:11.939 --> 00:03:14.460 talk about one specific feature, the outpainting 00:03:14.460 --> 00:03:17.259 fix. What is that? Oh, this is fantastic. Seriously 00:03:17.259 --> 00:03:19.259 practical. Let's say you have a great photo, 00:03:19.340 --> 00:03:21.699 maybe a portrait, but it's vertical. And you 00:03:21.699 --> 00:03:24.080 desperately need it as a horizontal banner, you 00:03:24.080 --> 00:03:27.080 know, 16 .9 aspect ratio for a website or YouTube. 00:03:27.300 --> 00:03:29.659 You just upload that vertical image to the free 00:03:29.659 --> 00:03:31.300 Gemini image tool. Then you prompted something 00:03:31.300 --> 00:03:34.580 like, expand the canvas. Reconstruct a realistic 00:03:34.580 --> 00:03:37.680 beach, sea, and sky to fill the new space. And 00:03:37.680 --> 00:03:40.159 the AI just does it. It fills in the missing 00:03:40.159 --> 00:03:42.020 background seamlessly. It keeps the perspective 00:03:42.020 --> 00:03:43.860 right. It doesn't mess up the faces or the pose. 00:03:44.060 --> 00:03:45.939 The whole thing looks photographically coherent. 00:03:46.060 --> 00:03:48.340 It's impressive. Or think about e -commerce. 00:03:48.539 --> 00:03:50.539 You've got a basic product photo, maybe on a 00:03:50.539 --> 00:03:52.819 white background, a PNG file. You can upload 00:03:52.819 --> 00:03:55.219 that and tell the AI, place this exact product 00:03:55.219 --> 00:03:57.740 in a high -end professional studio setting. Add 00:03:57.740 --> 00:04:00.060 realistic rim lighting and soft shadows. And 00:04:00.060 --> 00:04:02.870 boom. You get studio -quality shots. That kind 00:04:02.870 --> 00:04:05.189 of capability can save you thousands on actual 00:04:05.189 --> 00:04:07.669 photography setups. Wow. Okay, so that gives 00:04:07.669 --> 00:04:10.069 you the raw power to create and modify images. 00:04:10.370 --> 00:04:13.069 But what about consistency, especially for branding? 00:04:13.129 --> 00:04:15.370 That's where another tool comes in, right? Gemini 00:04:15.370 --> 00:04:17.230 Gems. You mentioned these. Tell us about Gems. 00:04:17.310 --> 00:04:20.290 Exactly. Gems are the key to consistency and 00:04:20.290 --> 00:04:23.779 efficiency for repetitive tasks. Think of gems 00:04:23.779 --> 00:04:26.480 as personalized AI agents. You train them on 00:04:26.480 --> 00:04:29.180 your specific needs, your style, your tone, or 00:04:29.180 --> 00:04:31.120 crucially for businesses, your actual products. 00:04:31.259 --> 00:04:33.279 They get optimized for those recurring jobs. 00:04:33.420 --> 00:04:35.379 And honestly, I still wrestle with prompt drift 00:04:35.379 --> 00:04:37.560 myself sometimes. You know, it happens. Okay, 00:04:37.600 --> 00:04:39.660 hang on. Prompt drift, what exactly is that? 00:04:40.480 --> 00:04:42.019 Prompt drift, it's when you start generating 00:04:42.019 --> 00:04:44.980 things with an AI, maybe a few images or some 00:04:44.980 --> 00:04:48.459 text. And after a few tries, the AI kind of forgets. 00:04:49.050 --> 00:04:51.430 Your original core instructions or your brand 00:04:51.430 --> 00:04:54.550 guidelines. It just starts wandering off. The 00:04:54.550 --> 00:04:56.230 results become inconsistent. It's frustrating. 00:04:56.949 --> 00:04:59.930 Gems are designed to solve exactly that. They 00:04:59.930 --> 00:05:01.610 use something called the knowledge field. This 00:05:01.610 --> 00:05:03.069 is where you upload your reference materials, 00:05:03.310 --> 00:05:05.569 your brand style guide, reference images of your 00:05:05.569 --> 00:05:07.490 product, technical specs, whatever. You put it 00:05:07.490 --> 00:05:09.310 right into that knowledge field for that specific 00:05:09.310 --> 00:05:13.509 gem. And the AI agent, it checks. those documents 00:05:13.509 --> 00:05:15.209 every single time it generates something for 00:05:15.209 --> 00:05:17.629 you okay the integration there sounds powerful 00:05:17.629 --> 00:05:19.990 so it's not just about making one image consistent 00:05:19.990 --> 00:05:23.069 it sounds like it enables scale how does that 00:05:23.069 --> 00:05:25.230 knowledge field really impact a business trying 00:05:25.230 --> 00:05:27.769 to produce a lot of content it enables massive 00:05:27.769 --> 00:05:30.730 scalability yeah because it locks in the critical 00:05:30.730 --> 00:05:33.209 details the products shape color proportions 00:05:33.209 --> 00:05:35.250 so you can generate maybe a thousand different 00:05:35.250 --> 00:05:37.589 images or variations and they all stick to the 00:05:37.589 --> 00:05:39.810 brand guidelines without you needing to manually 00:05:39.810 --> 00:05:41.930 check every single one that's huge All right, 00:05:41.949 --> 00:05:44.509 let's shift gears a bit from creating visuals 00:05:44.509 --> 00:05:48.829 to research and learning. Let's talk about Notebook 00:05:48.829 --> 00:05:51.129 LM. You mentioned this isn't like a regular chat 00:05:51.129 --> 00:05:52.889 bot browsing the whole web. It's more specialized. 00:05:53.230 --> 00:05:55.410 That's right. It's a specialized AI research 00:05:55.410 --> 00:05:58.149 assistant. Very different. Its whole purpose 00:05:58.149 --> 00:06:00.490 is to talk only to the documents you provide. 00:06:00.730 --> 00:06:03.269 You create a notebook and you fill it with your 00:06:03.269 --> 00:06:06.110 source materials, PDFs, Google Docs, articles 00:06:06.110 --> 00:06:08.470 from the web, even transcripts from YouTube videos. 00:06:08.649 --> 00:06:10.730 And when Notebook LM answers your questions, 00:06:10.910 --> 00:06:14.089 the answers are grounded. Grounded. OK, define 00:06:14.089 --> 00:06:16.370 that for us simply. Grounded just means its answers 00:06:16.370 --> 00:06:19.529 are strictly based on and sourced from the specific 00:06:19.529 --> 00:06:22.939 documents you uploaded. Nothing else. In fact, 00:06:23.019 --> 00:06:25.319 it usually highlights the exact passage in your 00:06:25.319 --> 00:06:28.279 source material that supports its claim, which 00:06:28.279 --> 00:06:30.560 makes the responses incredibly reliable. Great 00:06:30.560 --> 00:06:32.699 for serious research, whether it's academic or 00:06:32.699 --> 00:06:35.139 for business strategy. And the privacy angle 00:06:35.139 --> 00:06:38.110 here is really important, too. Notebook LM. acts 00:06:38.110 --> 00:06:39.810 like a private sandbox for your information. 00:06:40.110 --> 00:06:42.110 So if you're working on, say, a confidential 00:06:42.110 --> 00:06:44.689 new product idea or analyzing sensitive competitor 00:06:44.689 --> 00:06:47.389 data, you can upload those proprietary documents 00:06:47.389 --> 00:06:49.730 and ask questions about them without worrying 00:06:49.730 --> 00:06:52.290 that your data is going to be used to train some 00:06:52.290 --> 00:06:55.230 public AI model. That sounds fantastic, but people 00:06:55.230 --> 00:06:57.509 are always concerned about privacy with these 00:06:57.509 --> 00:06:59.810 tools. If I'm uploading really sensitive stuff, 00:07:00.009 --> 00:07:02.470 business plans, unpublished research, how confident 00:07:02.470 --> 00:07:04.910 can I actually be that it stays private? Yeah, 00:07:04.949 --> 00:07:06.990 that's a fair question. The confidence level 00:07:06.990 --> 00:07:09.509 should be pretty high. Because Google explicitly 00:07:09.509 --> 00:07:13.810 positions this tool as a private workspace. Their 00:07:13.810 --> 00:07:16.230 documentation states that your uploaded documents 00:07:16.230 --> 00:07:18.569 remain private to you, they are not used for 00:07:18.569 --> 00:07:21.509 ad personalization, and crucially, not used to 00:07:21.509 --> 00:07:24.490 train the public versions of Gemini. It really 00:07:24.490 --> 00:07:27.129 is designed to be a secure sandbox for your sensitive 00:07:27.129 --> 00:07:29.329 thinking and analysis. And then you can layer 00:07:29.329 --> 00:07:31.490 on Gemini Live. This is where things get really 00:07:31.490 --> 00:07:33.629 interactive. It basically turns the chatbot experience 00:07:33.629 --> 00:07:37.670 into a, well, a multimodal assistant. it permission 00:07:37.670 --> 00:07:40.209 it can access your screen you can hear your voice 00:07:40.209 --> 00:07:42.009 through the light see through your webcam it 00:07:42.009 --> 00:07:43.850 makes the whole conversation fully contextual 00:07:43.850 --> 00:07:45.689 it knows what you're looking at and talking about 00:07:45.689 --> 00:07:47.649 right and the example you shared about the ai 00:07:47.649 --> 00:07:49.990 design coach was pretty compelling can you walk 00:07:49.990 --> 00:07:52.129 us through that someone was working on a youtube 00:07:52.129 --> 00:07:54.329 thumbnail yeah exactly a creator shared their 00:07:54.329 --> 00:07:56.889 screen showing a thumbnail data design and they 00:07:56.889 --> 00:07:58.370 just asked gemini live you know how can i make 00:07:58.370 --> 00:08:01.250 this better get more clicks and because gemini 00:08:01.250 --> 00:08:04.180 live could see the thumbnail It analyzed the 00:08:04.180 --> 00:08:07.019 composition right there in real time. It suggested 00:08:07.019 --> 00:08:09.680 things like increasing the color contrast. Why? 00:08:09.860 --> 00:08:12.360 For better visibility on small mobile screens. 00:08:12.660 --> 00:08:15.860 Makes sense. It even suggested shifting where 00:08:15.860 --> 00:08:18.279 the text was placed so the viewer's eye would 00:08:18.279 --> 00:08:20.839 catch the main question first. That's not just 00:08:20.839 --> 00:08:23.660 generic advice. That's specific, actionable feedback 00:08:23.660 --> 00:08:25.699 happening instantly. Exactly. It's that real 00:08:25.699 --> 00:08:27.939 -time loop that's powerful. Okay, looking ahead 00:08:27.939 --> 00:08:31.579 slightly. Gemini 2 .5 computer. This isn't fully 00:08:31.579 --> 00:08:33.639 rolled out everywhere yet, but the vision is. 00:08:33.720 --> 00:08:35.799 Well, it's true delegation. We're talking about 00:08:35.799 --> 00:08:37.820 AI that can actually operate your computer for 00:08:37.820 --> 00:08:39.940 you. Browse websites, click buttons, fill out 00:08:39.940 --> 00:08:43.179 forms. So moving beyond just summarizing information 00:08:43.179 --> 00:08:46.220 or generating content, the AI could actually 00:08:46.220 --> 00:08:49.679 do complex tasks, multi -step things. Like what 00:08:49.679 --> 00:08:51.960 kind of future automation are we envisioning 00:08:51.960 --> 00:08:54.460 here? Think about comparative research. Instead 00:08:54.460 --> 00:08:56.620 of you manually going to 10 competitor websites, 00:08:57.000 --> 00:08:59.120 copying pricing data, putting it in a spreadsheet, 00:08:59.299 --> 00:09:01.779 you could potentially ask the AI, go to these 00:09:01.779 --> 00:09:04.399 10 sites, gather all their current pricing tiers 00:09:04.399 --> 00:09:06.519 for product X, and create a comparison table 00:09:06.519 --> 00:09:09.659 for me, highlighting the key differences. That 00:09:09.659 --> 00:09:12.539 level of automation, data gathering, complex 00:09:12.539 --> 00:09:14.919 form filling, system interactions, that's the 00:09:14.919 --> 00:09:16.919 direction. And we shouldn't forget the free voice 00:09:16.919 --> 00:09:18.559 generator either. This thing is surprisingly 00:09:18.559 --> 00:09:20.759 good. It really does rival some of those platforms 00:09:20.759 --> 00:09:23.879 that charge, you know, $20, $30, even $50 a month. 00:09:24.019 --> 00:09:26.159 You can configure the voice dial, conversational, 00:09:26.200 --> 00:09:28.320 authoritative, whatever you need. Pick an accent, 00:09:28.419 --> 00:09:30.620 adjust the pace. And crucially, it handles long 00:09:30.620 --> 00:09:32.679 form content pretty well. It can generate up 00:09:32.679 --> 00:09:36.330 to about eight minutes of audio. from 10 ,000 00:09:36.330 --> 00:09:38.690 characters of text in one go. That really is 00:09:38.690 --> 00:09:41.090 a game changer, especially for creators on a 00:09:41.090 --> 00:09:43.350 budget, people needing professional sounding 00:09:43.350 --> 00:09:46.090 narration for YouTube videos, or maybe creating 00:09:46.090 --> 00:09:49.029 e -learning modules at scale without hiring voice 00:09:49.029 --> 00:09:51.549 actors for every single piece. For sure. Which 00:09:51.549 --> 00:09:53.889 brings us neatly to the strategic part, because 00:09:53.889 --> 00:09:56.009 just having these tools isn't enough, right? 00:09:56.090 --> 00:09:58.370 You need to use them smartly. We've distilled 00:09:58.370 --> 00:10:01.870 it down to about six key rules for really maximizing 00:10:01.870 --> 00:10:03.909 the value you get from these free Google tools. 00:10:04.679 --> 00:10:07.159 First one is maybe the most important. Focus 00:10:07.159 --> 00:10:10.340 on workflows, not just individual tools. See, 00:10:10.419 --> 00:10:12.940 a single tool solves a single problem like generating 00:10:12.940 --> 00:10:15.799 an image, but a workflow solves a bottleneck 00:10:15.799 --> 00:10:18.200 in your whole process. Okay, workflow over tools. 00:10:18.340 --> 00:10:20.700 Makes sense. What's rule number two? Rule two 00:10:20.700 --> 00:10:23.500 is build reusable systems and templates. Don't 00:10:23.500 --> 00:10:25.980 just do one -off tasks. Invest a little time 00:10:25.980 --> 00:10:28.480 up front to create, say, a specialized Gemini 00:10:28.480 --> 00:10:31.580 gem for generating your social media posts. Train 00:10:31.580 --> 00:10:33.740 it on your brand voice, your preferred post structure, 00:10:33.940 --> 00:10:36.059 the kind of questions you like to ask. Then reuse 00:10:36.059 --> 00:10:39.120 that gem over and over. Efficiency. I like that. 00:10:39.220 --> 00:10:41.580 Build once, use many times. What's next? Rule 00:10:41.580 --> 00:10:44.320 three is about cross -tool synergy. This is where 00:10:44.320 --> 00:10:46.559 it gets really powerful. Don't use these tools 00:10:46.559 --> 00:10:49.159 in isolation. Think about it. Take the detailed, 00:10:49.240 --> 00:10:51.940 super reliable research you gathered using Notebook 00:10:51.940 --> 00:10:54.629 LM. And feed that specific information directly 00:10:54.629 --> 00:10:57.110 into your specialized Gemini gem when you ask 00:10:57.110 --> 00:10:59.129 it to write something. You're combining the grounded 00:10:59.129 --> 00:11:01.970 accuracy of Notebook LM with the stylistic consistency 00:11:01.970 --> 00:11:05.110 of gems. That's stacking the Lego blocks, you 00:11:05.110 --> 00:11:07.389 see. Ah, okay. Connecting the data flow between 00:11:07.389 --> 00:11:10.500 the tools. Got it. Rule four. Rule four. Embrace 00:11:10.500 --> 00:11:12.759 real -time iteration. This ties back to Gemini 00:11:12.759 --> 00:11:15.399 Live. Use that capability for instant feedback. 00:11:15.480 --> 00:11:18.039 Get the AI's suggestion on your design. Apply 00:11:18.039 --> 00:11:20.000 the change immediately. Maybe run it by the AI 00:11:20.000 --> 00:11:22.820 again. It dramatically speeds up that quality 00:11:22.820 --> 00:11:25.860 control and refinement process. Don't wait. Iterate 00:11:25.860 --> 00:11:28.080 fast. Makes sense. Get feedback. Tweak. Repeat. 00:11:28.279 --> 00:11:31.320 What's rule five? Rule five is batch processing. 00:11:31.620 --> 00:11:33.860 This is a practical one. Remember, these are 00:11:33.860 --> 00:11:36.840 free tools. They will likely have usage limits, 00:11:36.980 --> 00:11:39.720 daily quotas, things like that. So if you know 00:11:39.720 --> 00:11:42.340 you need, say, 50 slightly different product 00:11:42.340 --> 00:11:45.000 images for your website or maybe 10 minutes of 00:11:45.000 --> 00:11:47.440 voice narration for a video, don't do them one 00:11:47.440 --> 00:11:50.419 by one over several days. Plan ahead. Generate 00:11:50.419 --> 00:11:53.200 them all in one dedicated session. Chunking your 00:11:53.200 --> 00:11:55.580 tasks like this helps you avoid hitting those 00:11:55.580 --> 00:11:58.000 frustrating rate limits right when you're facing 00:11:58.000 --> 00:12:00.539 a deadline. That's a really practical tip. batching 00:12:00.539 --> 00:12:03.080 work. But speaking of limits, what are the realistic 00:12:03.080 --> 00:12:05.500 constraints we should be aware of? Are the rate 00:12:05.500 --> 00:12:07.840 limits restrictive? That leads directly into 00:12:07.840 --> 00:12:10.700 rule six. Know and optimize your rate limits. 00:12:10.860 --> 00:12:13.320 You absolutely need to understand what the daily 00:12:13.320 --> 00:12:16.200 or hourly boundaries are for each tool. If, for 00:12:16.200 --> 00:12:18.480 example, the free Gemini image tool allows you 00:12:18.480 --> 00:12:21.139 50 generations per day, don't waste 10 of those 00:12:21.139 --> 00:12:23.299 just experimenting randomly to find a new visual 00:12:23.299 --> 00:12:25.789 style you might like. Instead, use a gem you've 00:12:25.789 --> 00:12:28.149 already refined, rule two. Batch process the 00:12:28.149 --> 00:12:30.389 40 final images you actually need, rule five, 00:12:30.529 --> 00:12:32.850 and save the open -ended experimentation for 00:12:32.850 --> 00:12:35.230 another day when you have quota to spare. Be 00:12:35.230 --> 00:12:37.370 strategic about it. So just to reiterate that 00:12:37.370 --> 00:12:40.100 first point, because it seems foundational. Why 00:12:40.100 --> 00:12:43.860 is it so important for users to prioritize building 00:12:43.860 --> 00:12:46.200 workflows rather than just collecting a bunch 00:12:46.200 --> 00:12:48.679 of different tools? Because workflows solve the 00:12:48.679 --> 00:12:51.039 real problems, the bottlenecks in getting things 00:12:51.039 --> 00:12:53.580 done efficiently and consistently. Individual 00:12:53.580 --> 00:12:56.720 tools are just components. A good workflow connects 00:12:56.720 --> 00:12:59.100 them into a system that actually multiplies their 00:12:59.100 --> 00:13:01.039 value. It's about the whole system, not just 00:13:01.039 --> 00:13:04.500 the parts. Midroll sponsor read. OK, we've covered 00:13:04.500 --> 00:13:07.480 the core free tools and how to use them strategically, 00:13:07.759 --> 00:13:11.039 but we should also touch on the the wider Google 00:13:11.039 --> 00:13:14.360 ecosystem here. There's Google Labs, right? What's 00:13:14.360 --> 00:13:16.940 that about? Yeah. Google Labs is essentially 00:13:16.940 --> 00:13:19.120 their experimental playground. It's where they 00:13:19.120 --> 00:13:21.340 put out early stage projects for public testing 00:13:21.340 --> 00:13:23.220 and feedback. It's kind of like getting a peek 00:13:23.220 --> 00:13:25.840 inside their R &D workshop. You can often find 00:13:25.840 --> 00:13:27.639 really interesting things there. For instance, 00:13:27.740 --> 00:13:29.899 you might get early access to tools like Music 00:13:29.899 --> 00:13:32.340 FX, which generates instrumental music tracks 00:13:32.340 --> 00:13:35.039 based on prompts. Or sometimes they release early 00:13:35.039 --> 00:13:37.539 versions of more advanced capabilities, like 00:13:37.539 --> 00:13:39.559 their video generation model, VO, for people 00:13:39.559 --> 00:13:42.299 to try out. It's the bleeding edge. And thinking 00:13:42.299 --> 00:13:44.779 about all these tools connecting, Notebook LM 00:13:44.779 --> 00:13:47.600 feeding gems, Gemini Live providing feedback, 00:13:47.840 --> 00:13:51.980 the scale is just... Whoa. I mean, imagine scaling 00:13:51.980 --> 00:13:54.000 workflows built on these principles to handle, 00:13:54.100 --> 00:13:57.399 I don't know, a billion queries or tasks. The 00:13:57.399 --> 00:13:59.399 potential efficiency is kind of staggering. It 00:13:59.399 --> 00:14:02.019 really is. Which brings us back to that fundamental 00:14:02.019 --> 00:14:05.600 question. Why? Why is Google giving away tools 00:14:05.600 --> 00:14:08.399 that are clearly powerful and frankly must be 00:14:08.399 --> 00:14:10.899 expensive to develop and run? It's definitely 00:14:10.899 --> 00:14:12.919 not charity, right? It's smart business strategy. 00:14:13.000 --> 00:14:15.139 Yeah. And it seems to rest on about three main 00:14:15.139 --> 00:14:17.179 pillars. Okay, let's hear them. What's pillar 00:14:17.179 --> 00:14:20.169 number one? Ecosystem lock -in. It's a classic 00:14:20.169 --> 00:14:22.269 strategy. If you build your entire creative or 00:14:22.269 --> 00:14:24.309 business workflow around Google's free tools, 00:14:24.470 --> 00:14:26.610 if all your research data is in Notebook LM, 00:14:26.809 --> 00:14:29.250 your custom agents are built as gems, your visuals 00:14:29.250 --> 00:14:31.710 come from Gemini Image, the cost and hassle of 00:14:31.710 --> 00:14:34.070 switching to a competitor later on, like Microsoft's 00:14:34.070 --> 00:14:37.110 Copilot or tools from Anthropic or OpenAI, becomes 00:14:37.110 --> 00:14:39.769 really high. It creates a powerful moat. You're 00:14:39.769 --> 00:14:41.450 just less likely to leave the ecosystem you're 00:14:41.450 --> 00:14:43.870 already deeply integrated into. Makes perfect 00:14:43.870 --> 00:14:46.820 sense. Pillar number two. Data and improvement. 00:14:47.320 --> 00:14:50.120 Even if the data is aggregated and anonymized, 00:14:50.240 --> 00:14:52.559 every interaction you have with these free tools 00:14:52.559 --> 00:14:55.019 provides Google with valuable real -world usage 00:14:55.019 --> 00:14:57.740 data. They see which prompts work well, which 00:14:57.740 --> 00:15:00.220 types of tasks users struggle with, where the 00:15:00.220 --> 00:15:03.139 tools fail or cause frustration. It's essentially 00:15:03.139 --> 00:15:05.799 free research and development for them. It helps 00:15:05.799 --> 00:15:08.100 them understand user behavior at scale and rapidly 00:15:08.100 --> 00:15:10.860 improve their underlying AI models. Okay, lock 00:15:10.860 --> 00:15:14.440 -in, free R &D. What's the third pillar? It's 00:15:14.440 --> 00:15:17.179 the gateway to paid services, the classic freemium 00:15:17.179 --> 00:15:19.919 model. By offering compelling free tools for 00:15:19.919 --> 00:15:22.620 individuals and small teams, Google builds familiarity, 00:15:22.919 --> 00:15:25.850 trust, and dependence on their ecosystem. Then, 00:15:25.950 --> 00:15:27.909 when your business grows or you need more advanced 00:15:27.909 --> 00:15:30.330 enterprise features, maybe higher usage limits, 00:15:30.490 --> 00:15:33.029 enhanced security, dedicated support, or access 00:15:33.029 --> 00:15:35.309 to their high -powered Google Cloud AI services, 00:15:35.690 --> 00:15:37.669 you're naturally going to gravitate towards the 00:15:37.669 --> 00:15:39.490 paid offerings within the Google ecosystem you 00:15:39.490 --> 00:15:41.629 already know and use. It converts free users 00:15:41.629 --> 00:15:43.750 into high -margin enterprise customers down the 00:15:43.750 --> 00:15:46.789 line. Now, a crucial warning for everyone listening 00:15:46.789 --> 00:15:48.909 who's getting excited about this. While these 00:15:48.909 --> 00:15:51.450 tools are free now, it's wise to assume that 00:15:51.450 --> 00:15:53.990 this level of unrestricted free access might 00:15:53.990 --> 00:15:56.470 be temporary. or at least that certain features 00:15:56.470 --> 00:15:59.429 will evolve. We should probably expect stricter 00:15:59.429 --> 00:16:02.129 usage limits over time or perhaps see the most 00:16:02.129 --> 00:16:04.610 advanced capabilities migrating into paid tiers 00:16:04.610 --> 00:16:07.750 eventually. So the message is, use this free 00:16:07.750 --> 00:16:10.330 access now. Build your skills, develop your workflows, 00:16:10.570 --> 00:16:12.710 gain that institutional knowledge while the gates 00:16:12.710 --> 00:16:15.740 are wide open. So boiling it down, if these tools 00:16:15.740 --> 00:16:18.580 are genuinely so good and powerful, what's the 00:16:18.580 --> 00:16:20.820 single biggest business reason Google is just 00:16:20.820 --> 00:16:22.980 giving them away right now? I'd say it's primarily 00:16:22.980 --> 00:16:25.659 to establish and solidify that ecosystem lock 00:16:25.659 --> 00:16:28.799 -in, to build a strong, defensible moat against 00:16:28.799 --> 00:16:31.279 their major competitors like Microsoft and OpenAI, 00:16:31.559 --> 00:16:34.980 ensuring users build their AI habits within Google's 00:16:34.980 --> 00:16:38.669 world. Okay, so let's wrap things up. The big 00:16:38.669 --> 00:16:40.950 idea here is that these free Google AI tools 00:16:40.950 --> 00:16:43.730 represent a genuinely powerful new productivity 00:16:43.730 --> 00:16:46.389 arsenal for you. But remember, they're designed 00:16:46.389 --> 00:16:48.429 to augment your intelligence and creativity, 00:16:48.629 --> 00:16:51.429 not replace it entirely. The real magic happens 00:16:51.429 --> 00:16:53.389 when you combine them strategically, like we 00:16:53.389 --> 00:16:56.029 discussed, using Notebook LM for deep, grounded 00:16:56.029 --> 00:16:58.350 research, and then feeding those specific insights 00:16:58.350 --> 00:17:00.730 into a custom Gemini gem to generate content 00:17:00.730 --> 00:17:02.909 that's both accurate and perfectly on brand. 00:17:03.029 --> 00:17:05.369 That's the synergy. And one final really important 00:17:05.369 --> 00:17:07.890 security reminder. Please always use the official 00:17:07.890 --> 00:17:10.049 Google platforms for these tools. Go directly 00:17:10.049 --> 00:17:12.849 to astudio .google .com for the image and gem 00:17:12.849 --> 00:17:15.849 tools or notebookalum .google .com for the research 00:17:15.849 --> 00:17:17.910 assistant. We're already seeing scams pop up 00:17:17.910 --> 00:17:19.750 online, websites charging people for special 00:17:19.750 --> 00:17:22.029 access or premium versions of tools that Google 00:17:22.029 --> 00:17:23.849 is actually providing for free. Don't fall for 00:17:23.849 --> 00:17:26.089 those. Stick to the official sources. Well said. 00:17:26.349 --> 00:17:29.369 So the age of truly accessible, powerful AI is 00:17:29.369 --> 00:17:31.809 definitely here. And right now, a significant 00:17:31.809 --> 00:17:34.880 part of it is free. Your mission? Should you 00:17:34.880 --> 00:17:37.190 choose to accept it this week? Pick just one 00:17:37.190 --> 00:17:39.049 of the tools we talked about today, maybe Gemini 00:17:39.049 --> 00:17:42.470 Image or Gems or Notebook LM, and dedicate just 00:17:42.470 --> 00:17:45.650 15 minutes. That's it. 15 minutes to exploring 00:17:45.650 --> 00:17:48.089 and maybe mastering one specific feature or strategic 00:17:48.089 --> 00:17:50.410 nuance we discussed, like creating your first 00:17:50.410 --> 00:17:53.049 knowledge field in a gem or trying the outpainting 00:17:53.049 --> 00:17:56.250 feature. Just take one small step. And finally, 00:17:56.269 --> 00:17:58.450 here's something to chew on after you're done 00:17:58.450 --> 00:18:01.609 listening. If powerful free AI like this becomes 00:18:01.609 --> 00:18:03.730 the sort of default plumbing underlying almost 00:18:03.730 --> 00:18:05.990 all digital products and creative workflows, 00:18:06.549 --> 00:18:08.789 What fundamental human skill do you think suddenly 00:18:08.789 --> 00:18:10.789 becomes the most valuable bottleneck? What human 00:18:10.789 --> 00:18:13.549 ability becomes the premium? Beat! Something 00:18:13.549 --> 00:18:14.150 to think about.