WEBVTT 00:00:00.000 --> 00:00:02.520 The biggest shift happening in AI right now isn't 00:00:02.520 --> 00:00:05.259 about, you know, better chatbots that talk more 00:00:05.259 --> 00:00:07.879 smoothly. No. It's about making them do things. 00:00:08.039 --> 00:00:11.660 It's this move from informational AI to what 00:00:11.660 --> 00:00:14.519 we're calling agentic AI. Exactly. I mean, imagine 00:00:14.519 --> 00:00:17.019 an AI agent that doesn't just give you a product 00:00:17.019 --> 00:00:21.579 suggestion, but actually manages the entire transaction. 00:00:21.879 --> 00:00:24.839 It's applying live discounts, confirming your 00:00:24.839 --> 00:00:27.039 address, handling the payment, and submitting 00:00:27.039 --> 00:00:29.850 the order. All of it. And that world, it seems, 00:00:30.010 --> 00:00:32.609 just arrived. It's here. Welcome to the Deep 00:00:32.609 --> 00:00:35.509 Dive. We've synthesized a massive stack of sources 00:00:35.509 --> 00:00:38.009 today, and they're all focused on this acceleration 00:00:38.009 --> 00:00:41.250 toward agentic AI. Which really is just a simple 00:00:41.250 --> 00:00:44.049 concept. It's AI that takes actions on your behalf. 00:00:44.369 --> 00:00:47.250 So today we're unpacking three major themes that 00:00:47.250 --> 00:00:49.750 show where this is all going. Yep, and we're 00:00:49.750 --> 00:00:51.250 starting with a huge commercial announcement, 00:00:51.509 --> 00:00:53.969 the debut of agentic commerce, and the new standards 00:00:53.969 --> 00:00:56.390 that allow an AI to manage that entire shopping 00:00:56.390 --> 00:00:58.859 flow. Then second, we'll take a critical status 00:00:58.859 --> 00:01:01.380 check on the broader AI landscape. Yeah, we're 00:01:01.380 --> 00:01:03.219 going to balance the market hype with some very 00:01:03.219 --> 00:01:05.579 real, very high stakes medical applications. 00:01:05.879 --> 00:01:08.379 And we'll also dive into how Google is changing 00:01:08.379 --> 00:01:11.640 the rules for creating online content. And finally, 00:01:11.640 --> 00:01:13.519 we're going to explore a verified mathematical 00:01:13.519 --> 00:01:16.760 breakthrough. This one is amazing. An AI solved 00:01:16.760 --> 00:01:20.599 a decades old unsolved math problem with proofs 00:01:20.599 --> 00:01:22.640 that, you know, leading experts have confirmed 00:01:22.640 --> 00:01:26.859 are truly original. this changes things it really 00:01:26.859 --> 00:01:29.560 does okay it's a huge amount of ground to cover 00:01:29.560 --> 00:01:32.000 let's do it let's unpack this and start with 00:01:32.000 --> 00:01:35.620 that commerce revolution okay so starting right 00:01:35.620 --> 00:01:38.819 where the rubber hits the digital road commerce 00:01:38.819 --> 00:01:42.400 at the 2026 national retail federation conference 00:01:42.400 --> 00:01:45.739 the nrf google launched something pretty monumental 00:01:45.739 --> 00:01:48.359 it's called the universal commerce protocol or 00:01:48.359 --> 00:01:52.049 ucp what's so fascinating here is that ucp It's 00:01:52.049 --> 00:01:53.930 not just another Google feature. It's an open 00:01:53.930 --> 00:01:55.489 standard. Okay. What does that mean exactly? 00:01:55.769 --> 00:01:57.950 Think of it like a set of rules, you know, a 00:01:57.950 --> 00:02:00.049 common language that's designed specifically 00:02:00.049 --> 00:02:04.150 to let AI agents, any agent, really execute the 00:02:04.150 --> 00:02:06.629 entire commerce flow. So not just Google's agent. 00:02:06.870 --> 00:02:09.289 Exactly. Across all kinds of different platforms. 00:02:09.629 --> 00:02:11.969 This is basically the foundational plumbing for 00:02:11.969 --> 00:02:14.110 the next era of e -commerce. It's built for delegation. 00:02:14.430 --> 00:02:16.849 And the scale is what makes it instantly significant, 00:02:17.090 --> 00:02:19.389 right? This wasn't built in a vacuum. Not at 00:02:19.389 --> 00:02:22.020 all. Google partnered with giants right out of 00:02:22.020 --> 00:02:25.419 the gate. We're talking Shopify, Walmart, Etsy, 00:02:25.419 --> 00:02:29.240 Target, Wayfair. When those names sign on. That 00:02:29.240 --> 00:02:31.039 tells you the infrastructure for this is already 00:02:31.039 --> 00:02:33.639 massive. That partner list shows that the biggest 00:02:33.639 --> 00:02:36.900 players, they recognize that friction is what 00:02:36.900 --> 00:02:40.000 kills sales. Oh, absolutely. Their goal is radical 00:02:40.000 --> 00:02:42.280 simplicity. And where does this simplicity happen 00:02:42.280 --> 00:02:44.740 for the user? Inside the tools you're already 00:02:44.740 --> 00:02:48.240 using, like Gemini AI or Google Search's AI mode. 00:02:48.460 --> 00:02:50.300 Right. So instead of clicking through a bunch 00:02:50.300 --> 00:02:53.000 of links, reading reviews, adding to a cart, 00:02:53.159 --> 00:02:55.199 filling out four checkout pages. Yeah, all that 00:02:55.199 --> 00:02:57.719 hassle. Now, if you're searching for a U .S.-based 00:02:57.719 --> 00:02:59.680 product, you can... can see the listing and complete 00:02:59.680 --> 00:03:01.939 the whole transaction right from the search results 00:03:01.939 --> 00:03:05.520 window. It's end -to -end automation. And this 00:03:05.520 --> 00:03:07.439 is where the system gets really smart. It just 00:03:07.439 --> 00:03:10.020 integrates with payment methods you already trust. 00:03:10.460 --> 00:03:13.300 Google Pay or PayPal. Yep. And your shipping 00:03:13.300 --> 00:03:16.240 info is autofilled using the data that's already 00:03:16.240 --> 00:03:18.979 in your Google wallet. So it feels less like 00:03:18.979 --> 00:03:21.419 you're navigating a store and more like you're 00:03:21.419 --> 00:03:23.240 just confirming a decision you've already made. 00:03:23.340 --> 00:03:25.479 Which is nice. But I think the real power up 00:03:25.479 --> 00:03:28.560 here is actually for the merchants. How so? Well, 00:03:28.620 --> 00:03:31.400 before, a customer service agent might give you 00:03:31.400 --> 00:03:34.300 a static coupon code, right? Now, brands can 00:03:34.300 --> 00:03:37.060 program their systems to trigger live discounts 00:03:37.060 --> 00:03:40.849 during the AI shopping chat based... on real 00:03:40.849 --> 00:03:44.210 -time inventory or context. That's a huge operational 00:03:44.210 --> 00:03:47.030 shift. And it goes further. Merchants can now 00:03:47.030 --> 00:03:49.789 embed their own specialized AI -powered business 00:03:49.789 --> 00:03:52.610 agents right inside Google Search. So if I have 00:03:52.610 --> 00:03:54.889 a really complex question, something super specific. 00:03:55.150 --> 00:03:57.870 Like, does this solar panel kit meet the voltage 00:03:57.870 --> 00:04:00.270 requirements for my 1990 Airstream model? Right. 00:04:00.349 --> 00:04:02.330 The agent answers, and then you just check out. 00:04:02.389 --> 00:04:04.490 It cuts out so many layers of customer service. 00:04:04.590 --> 00:04:06.629 And to make all this work, Google had to upgrade 00:04:06.629 --> 00:04:09.110 their Merchant Center. They added new data attributes. 00:04:09.550 --> 00:04:12.310 So sellers need to provide much more precise, 00:04:12.530 --> 00:04:15.969 more structured data. than ever before. They're 00:04:15.969 --> 00:04:18.290 essentially prepping their products to be understood 00:04:18.290 --> 00:04:22.670 by these new conversational AI results. And the 00:04:22.670 --> 00:04:25.110 industry confirmed this was the direction immediately. 00:04:25.670 --> 00:04:29.689 We saw Shopify drop its own AI checkout integration 00:04:29.689 --> 00:04:32.689 with Microsoft Copilot the very same day. So 00:04:32.689 --> 00:04:35.370 they're all in on this unified standard. Clearly. 00:04:35.829 --> 00:04:39.029 The trend is unambiguous. E -commerce and health 00:04:39.029 --> 00:04:41.209 care, those are the two sectors driving this 00:04:41.209 --> 00:04:43.370 aggressive eugenic AI integration right now. 00:04:43.470 --> 00:04:46.329 The value is just so immediate. So we've laid 00:04:46.329 --> 00:04:49.980 out the tech and the partners. How fundamentally 00:04:49.980 --> 00:04:52.480 does the universal commerce protocol change the 00:04:52.480 --> 00:04:56.279 consumer habit of browsing for products? It instantly 00:04:56.279 --> 00:04:58.980 shifts shopping from navigating links to automated 00:04:58.980 --> 00:05:01.420 conversational transaction. That's a huge change. 00:05:01.699 --> 00:05:03.800 It is. And that transition from navigating to 00:05:03.800 --> 00:05:05.660 delegating, it brings us right to our status 00:05:05.660 --> 00:05:08.000 check on the broader AI landscape. Which is a 00:05:08.000 --> 00:05:10.000 field that feels like it's speeding up while 00:05:10.000 --> 00:05:13.040 also sort of fighting gravity. That's a good 00:05:13.040 --> 00:05:14.839 way to put it. The bubble question is still central. 00:05:15.000 --> 00:05:17.000 Right. Our sources compiled views from about 00:05:17.000 --> 00:05:19.269 40 tech leaders and analysts. on whether this 00:05:19.269 --> 00:05:22.370 is all just hype or a dip or a sustainable wave. 00:05:22.779 --> 00:05:24.939 And the consensus is. There is no consensus. 00:05:25.139 --> 00:05:27.540 The fact that there's such strong polarized disagreement 00:05:27.540 --> 00:05:30.220 tells you just how volatile the market sentiment 00:05:30.220 --> 00:05:32.779 still is. But the money is still flowing into 00:05:32.779 --> 00:05:35.220 practical applications, especially security. 00:05:35.680 --> 00:05:37.579 It's massive. Look at a company like Torque. 00:05:37.660 --> 00:05:41.220 They just raised $140 million, which puts their 00:05:41.220 --> 00:05:44.279 valuation at $1 .2 billion. And what's their 00:05:44.279 --> 00:05:47.639 focus? Their entire focus is AI -driven security 00:05:47.639 --> 00:05:51.240 operations. They're putting agentic AI right 00:05:51.240 --> 00:05:53.230 on the front lines. of cybersecurity, where the 00:05:53.230 --> 00:05:55.889 stakes are arguably the highest. Speaking of 00:05:55.889 --> 00:05:58.689 high stakes, healthcare remains the other critical 00:05:58.689 --> 00:06:02.350 frontier. Anthropic, following OpenAI's lead, 00:06:02.569 --> 00:06:05.670 just launched Clawed for Healthcare. And this 00:06:05.670 --> 00:06:07.949 is compelling for individual users. It allows 00:06:07.949 --> 00:06:10.089 you to link your Apple Health data directly to 00:06:10.089 --> 00:06:13.319 Clawed for analysis. So it's trying to democratize 00:06:13.319 --> 00:06:15.740 the understanding of your own complex medical 00:06:15.740 --> 00:06:18.339 data. Exactly. Giving you context for your own 00:06:18.339 --> 00:06:20.839 numbers. But the skepticism is and should be 00:06:20.839 --> 00:06:23.759 incredibly deep when AI touches diagnostics. 00:06:23.800 --> 00:06:26.579 Yeah. The CEO of Exelon Musk claimed that Grok 00:06:26.579 --> 00:06:29.240 could out -diagnose doctors. Which immediately 00:06:29.240 --> 00:06:31.759 triggered warnings from experts. They see this 00:06:31.759 --> 00:06:34.769 as just incredibly risky. I mean, I still wrestle 00:06:34.769 --> 00:06:37.009 with prompt drift myself. True. You know, when 00:06:37.009 --> 00:06:40.189 an LLM just slowly forgets the initial instructions 00:06:40.189 --> 00:06:43.389 and starts kind of wandering off topic. It gets 00:06:43.389 --> 00:06:45.730 lazy or starts hallucinating. That's a common 00:06:45.730 --> 00:06:48.350 issue. Yeah. And in health care, that drift or 00:06:48.350 --> 00:06:51.889 that inaccuracy can be fatal. Right. There was 00:06:51.889 --> 00:06:54.430 a stark example cited by experts where Grochmus 00:06:54.430 --> 00:06:57.290 took a breast cyst for testicles. during a diagnostic 00:06:57.290 --> 00:07:00.850 run. Wow. And even though Musk publicly shared 00:07:00.850 --> 00:07:04.110 his own MRI to demonstrate the tech, seeing these 00:07:04.110 --> 00:07:06.350 high -stakes errors makes it painfully clear 00:07:06.350 --> 00:07:10.009 how dangerous data inaccuracy is when we delegate 00:07:10.009 --> 00:07:12.610 critical decisions to a machine. It's the difference 00:07:12.610 --> 00:07:14.490 between buying the wrong sweater and getting 00:07:14.490 --> 00:07:17.149 the wrong diagnosis. The tolerance for error 00:07:17.149 --> 00:07:19.689 is basically zero. Absolutely. And that need 00:07:19.689 --> 00:07:21.810 for accuracy and quality is now fundamentally 00:07:21.810 --> 00:07:24.269 changing content creation too. You mean with 00:07:24.269 --> 00:07:26.269 Google? Yeah, Google just issued a significant 00:07:26.269 --> 00:07:28.290 warning to publishers. They told them to stop 00:07:28.290 --> 00:07:30.850 chunking content. What exactly does chunking 00:07:30.850 --> 00:07:33.589 mean here? It's breaking down complex articles 00:07:33.589 --> 00:07:36.970 into these tiny, thin, bite -sized summaries. 00:07:37.610 --> 00:07:40.550 Often really repetitive or low -quality stuff 00:07:40.550 --> 00:07:44.509 designed specifically to feed AI models or just 00:07:44.509 --> 00:07:47.329 rank quickly. It created a flood of low -effort 00:07:47.329 --> 00:07:49.569 information. Exactly. So this is a huge pivot. 00:07:49.649 --> 00:07:52.990 Google is actively rewarding high -quality, deep... 00:07:53.149 --> 00:07:55.569 human writing again. Pushing back against that 00:07:55.569 --> 00:07:59.910 flood of thin AI optimized summaries. Yep. They're 00:07:59.910 --> 00:08:01.790 saying if the content doesn't offer true value 00:08:01.790 --> 00:08:03.850 and depth. It's not what they're looking for 00:08:03.850 --> 00:08:05.870 anymore. And on the money side. The writing's 00:08:05.870 --> 00:08:08.129 on the wall. Sources confirm advertisements are 00:08:08.129 --> 00:08:10.310 definitely coming to AI mode within Google search. 00:08:10.569 --> 00:08:13.490 Those AI conversations are the new ad inventory. 00:08:13.750 --> 00:08:15.689 Before we transition, let's just touch on a few 00:08:15.689 --> 00:08:18.189 new tools showing some utility. Okay. Pain AI 00:08:18.189 --> 00:08:20.769 is one. It's focused on editing spreadsheet cells 00:08:20.769 --> 00:08:23.569 and formulas with what they claim is human -level 00:08:23.569 --> 00:08:26.029 precision. Which would be a massive time saver 00:08:26.029 --> 00:08:28.439 if it actually works. Then there's Lexing, which 00:08:28.439 --> 00:08:31.339 is kind of clever. It's a fully offline, Tamagotchi 00:08:31.339 --> 00:08:33.419 -style game that turns language practice into 00:08:33.419 --> 00:08:36.179 a daily habit. And Elser AI, which is focused 00:08:36.179 --> 00:08:39.240 on long -form video, promising consistent characters 00:08:39.240 --> 00:08:42.139 and a clear story from just one prompt. They're 00:08:42.139 --> 00:08:44.259 trying to solve that consistency problem. So 00:08:44.259 --> 00:08:47.519 if Google is now actively rewarding high -quality 00:08:47.519 --> 00:08:50.080 human writing again... What does that imply for 00:08:50.080 --> 00:08:52.539 content strategy going forward? Human depth and 00:08:52.539 --> 00:08:54.919 verifiable insight will now become more valuable 00:08:54.919 --> 00:08:57.879 than simply generating high volumes of AI -optimized 00:08:57.879 --> 00:09:00.259 summaries or surface -level noise. Okay, here's 00:09:00.259 --> 00:09:01.759 where the conversation gets really interesting. 00:09:02.120 --> 00:09:06.340 Diving into pure theoretical capability, we have 00:09:06.340 --> 00:09:10.500 confirmation that GPT -5 .2 Pro solved a legendary 00:09:10.500 --> 00:09:13.570 decades -old unsolved math problem. This is a 00:09:13.570 --> 00:09:15.730 monumental achievement, not just for the AI community, 00:09:15.850 --> 00:09:17.909 but for mathematics itself. Right. The specific 00:09:17.909 --> 00:09:21.169 problem was Erdos problem number 397, which is 00:09:21.169 --> 00:09:24.230 all about binomial coefficients. For decades, 00:09:24.389 --> 00:09:26.789 mathematicians have been completely stumped. 00:09:26.830 --> 00:09:29.690 And the kicker here, the proof wasn't just, you 00:09:29.690 --> 00:09:31.929 know, accepted by some AI enthusiast. It was 00:09:31.929 --> 00:09:34.870 verified by Terence Tao. Who is arguably the 00:09:34.870 --> 00:09:36.490 world's most respected living mathematician, 00:09:36.850 --> 00:09:40.230 a Fields medalist. He's the gold standard. And 00:09:40.230 --> 00:09:42.529 he confirmed the proof was absolutely valid. 00:09:42.990 --> 00:09:45.210 The verification chain itself is fascinating. 00:09:45.389 --> 00:09:47.610 It really highlights this collaboration between 00:09:47.610 --> 00:09:51.950 the AI's raw power and the formal discipline 00:09:51.950 --> 00:09:55.610 of math. It does. So Neil Samani prompted the 00:09:55.610 --> 00:09:59.309 AI with the problem. GPT then generated this 00:09:59.309 --> 00:10:01.559 complex solution. Then what happened? We saw 00:10:01.559 --> 00:10:03.620 these other components mentioned, like Aristotle 00:10:03.620 --> 00:10:06.039 and a language called Lean. What role did they 00:10:06.039 --> 00:10:09.299 play? So Aristotle acted as a logic checker and 00:10:09.299 --> 00:10:12.200 cleaner. It took the kind of messy natural language 00:10:12.200 --> 00:10:15.019 output from GPT and formalized the argument, 00:10:15.220 --> 00:10:18.240 stripping out any ambiguity. Then it converted 00:10:18.240 --> 00:10:20.570 that logic into Lean. which is a formal proof 00:10:20.570 --> 00:10:23.610 language. Think of lean as code that can be mathematically 00:10:23.610 --> 00:10:26.289 compiled and verified by a computer. It guarantees 00:10:26.289 --> 00:10:28.769 every single logical step is sound. And only 00:10:28.769 --> 00:10:31.549 then did Tower review the final proof. Exactly. 00:10:31.549 --> 00:10:34.190 That process, it confirms every step is sound, 00:10:34.250 --> 00:10:36.409 but it still begs that question of originality. 00:10:36.549 --> 00:10:39.230 Right. Back in 2025, there is that controversy 00:10:39.230 --> 00:10:42.990 that AI was only solving math by just resurfacing 00:10:42.990 --> 00:10:45.570 known solutions from its training data, like 00:10:45.570 --> 00:10:47.850 a really fast lookup engine. Precisely. That 00:10:47.850 --> 00:10:50.659 was the core doubt. But the crucial detail here 00:10:50.659 --> 00:10:53.639 is that Tao confirmed that these new proofs, 00:10:53.639 --> 00:10:56.539 not just for number 397, but for two others as 00:10:56.539 --> 00:10:59.779 well, are original. Not previously known. Not 00:10:59.779 --> 00:11:01.879 known or published by humans. Yeah. It truly 00:11:01.879 --> 00:11:04.659 generated a novel pathway to the solution. So 00:11:04.659 --> 00:11:07.179 it's not just a prodigy at memorization. It's 00:11:07.179 --> 00:11:10.279 a prodigy at synthesizing new verified knowledge. 00:11:10.559 --> 00:11:12.799 Yep. That makes three Erdos problems solved. 00:11:13.240 --> 00:11:16.200 with about 660 left to go in that set. But let's 00:11:16.200 --> 00:11:17.820 keep this grounded. It's not perfect, right? 00:11:17.879 --> 00:11:19.740 There are still major limitations. Oh, for sure. 00:11:19.820 --> 00:11:22.759 It proves exceptional synthesis, but GBT 5 .2 00:11:22.759 --> 00:11:25.059 still struggles gladly. We're talking only about 00:11:25.059 --> 00:11:28.700 a 25 % success rate. with more abstract, open 00:11:28.700 --> 00:11:31.720 -ended math that requires that deep, human -level 00:11:31.720 --> 00:11:33.860 conceptual insight. So it's great at finding 00:11:33.860 --> 00:11:36.000 pathways, but less great at making the initial 00:11:36.000 --> 00:11:39.159 abstract leap in totally unknown territory. But 00:11:39.159 --> 00:11:42.120 this ability to crack known bottlenecks, that 00:11:42.120 --> 00:11:45.159 is the true revolutionary power. It's like handing 00:11:45.159 --> 00:11:47.360 the machine a complicated knot that's resisted 00:11:47.360 --> 00:11:49.620 decades of human effort, and it just returns 00:11:49.620 --> 00:11:52.279 the sequence of moves to untie it. Yeah. Whoa. 00:11:53.000 --> 00:11:55.379 Imagine scaling that mathematical originality 00:11:55.379 --> 00:11:58.559 to a billion scientific queries or unsolved material 00:11:58.559 --> 00:12:00.799 science problems that are holding back innovation. 00:12:01.080 --> 00:12:04.379 That's it. We can now give the AI an Erdis -style 00:12:04.379 --> 00:12:07.440 bottleneck problem in any field, something that 00:12:07.440 --> 00:12:10.019 requires a confirmed breakthrough, not just an 00:12:10.019 --> 00:12:13.120 optimization. So does generating these original 00:12:13.120 --> 00:12:16.539 proofs mean the AI has attained true abstract 00:12:16.539 --> 00:12:19.700 mathematical insight? It proves exceptional synthesis 00:12:19.700 --> 00:12:22.440 and verification capability, yet deep abstract 00:12:22.440 --> 00:12:24.860 human intuition and insight remain the critical 00:12:24.860 --> 00:12:27.480 limiting factor. Okay, so today we tracked two 00:12:27.480 --> 00:12:30.200 massive shifts at the bleeding edge of AI capability. 00:12:30.559 --> 00:12:32.940 First, that decisive move from informational 00:12:32.940 --> 00:12:36.379 AI to agentic AI, which automates complex tasks, 00:12:36.580 --> 00:12:40.120 from commerce through UCP to high -stakes cybersecurity. 00:12:40.460 --> 00:12:42.200 This is the delegation of our routine decision 00:12:42.200 --> 00:12:45.019 -making. And second, the confirmation of original 00:12:45.019 --> 00:12:48.899 thought in AI. Specifically, GPT -5 .2's ability 00:12:48.899 --> 00:12:51.480 to solve these previously impenetrable mathematical 00:12:51.480 --> 00:12:54.659 logjams with novel proofs. And this marks the 00:12:54.659 --> 00:12:57.039 delegation of fundamental theoretical bottlenecks. 00:12:57.120 --> 00:12:59.580 The shift means we're delegating routine decisions 00:12:59.580 --> 00:13:02.159 to our new digital assistants, trusting them 00:13:02.159 --> 00:13:04.700 with our wallets and our time. Yeah. And the 00:13:04.700 --> 00:13:07.440 math breakthrough means we can now delegate fundamental 00:13:07.440 --> 00:13:10.980 scientific impasses, trusting them with the boundaries 00:13:10.980 --> 00:13:13.470 of human knowledge. And this raises a really 00:13:13.470 --> 00:13:15.429 important question for you, the listener, to 00:13:15.429 --> 00:13:19.570 consider. If AI can break decades -old scientific 00:13:19.570 --> 00:13:22.129 logjams in pure mathematics, which overlooked 00:13:22.129 --> 00:13:25.789 longstanding bottleneck problem in your own field, 00:13:25.830 --> 00:13:28.289 in your business, your research should you hand 00:13:28.289 --> 00:13:31.009 to an AI next? Something to mull over as you 00:13:31.009 --> 00:13:33.409 transition to a world where AI doesn't just inform 00:13:33.409 --> 00:13:35.990 you, but acts for you. Until the next deep dive, 00:13:36.210 --> 00:13:36.850 stay curious.