0.000000 2.720000 Tech It From Me is an independent and solo-produced podcast. 2.720000 6.120000 Welcome to the Tech It From Me podcast, I'm Mike Madole. 6.120000 10.160000 On today's episode, I will be talking about Nvidia after they became the world's first 10.160000 15.280000 $4 trillion publicly traded company on July 9th, 2025. 15.280000 19.320000 That's right, a company that started out making graphics cards for gamers is now worth 19.320000 21.880000 more than Amazon and Google combined. 21.880000 23.200000 How did this happen? 23.200000 26.200000 Why are Nvidia's chips so critical for AI? 26.200000 30.480000 Is this astronomical rise sustainable or are we headed into an AI bubble? 30.480000 34.880000 Let's unpack the Nvidia story when it means for you and where the future of AI hardware 34.880000 35.880000 is going. 35.880000 39.300000 This is Tech It From Me, let's go. 39.300000 41.480000 This is the Tech It From Me podcast. 41.480000 46.640000 To understand how Nvidia got here, let's rewind and head back to the early 1990s. 46.640000 53.400000 So back in 1993, Jensen Huang and two friends founded Nvidia in a small office in California. 53.400000 57.280000 Their bet was simple, the future of computing was visual. 57.280000 63.160000 At the time, CPUs handled everything, but they weren't optimized for complex graphics rendering. 63.160000 67.800000 Nvidia's breakthrough came in 1999 with the GeForce 256. 67.800000 72.840000 It was marketed as the world's first GPU or a graphics processing unit. 72.840000 80.000000 For gamers, this meant more realistic visuals, faster frame rates, and a better gaming experience. 80.000000 84.600000 Throughout the 2000s, Nvidia dominated the gaming GPU market. 84.600000 90.720000 Their brand became synonymous with high-end gaming machines, and competitors like ATI struggled 90.720000 91.960000 to keep up. 91.960000 95.200000 That is until AMD bought them. 95.200000 99.080000 But behind the scenes, something interesting was happening. 99.080000 103.120000 Researchers discovered that GPUs weren't just good for rendering graphics. 103.120000 108.280000 Their architecture, which was designed for massive parallel processing, made them ideal 108.280000 114.640000 for complex mathematical computations, like those needed for scientific simulations, 114.640000 118.640000 oil and gas exploration, and financial modeling. 118.640000 124.880000 This was Nvidia's first clue that their GPUs had applications far beyond gaming. 124.880000 129.880000 Moving forward a few years to 2006, Nvidia launched something that would change its trajectory 129.880000 131.360000 forever. 131.360000 132.360000 CUDA. 132.360000 137.320000 CUDA stands for Compute Unified Device Architecture. 137.320000 143.680000 For CUDA, GPUs were primarily used for graphics rendering, gaming visuals, 3D modeling, and 143.680000 145.480000 video processing. 145.480000 148.160000 But CUDA unlocked their true potential. 148.160000 153.800000 It allowed developers to program Nvidia GPUs for general purpose computing, meaning they 153.800000 160.120000 could now handle any workload that benefited from parallel processing, from scientific simulations 160.120000 162.360000 to financial modeling. 162.360000 166.400000 At the time, ATI was still a niche academic pursuit. 166.400000 170.900000 It confined to research labs with limited practical applications. 170.900000 176.800000 Neural networks existed, but training them on traditional CPUs was painfully slow and 176.800000 179.200000 very expensive. 179.200000 185.880000 Then came 2012, which was a landmark year for artificial intelligence. 185.880000 194.520000 Researchers led by Jeffrey Hinton used Nvidia GPUs to train AlexNet, a deep, convolutional, 194.520000 198.880000 neural network designed for computer vision. 198.880000 204.920000 AlexNet competed in the prestigious ImageNet competition, where it dramatically outperformed 204.920000 211.720000 all previous models, slashing the top five error rate by a humongous margin. 211.720000 214.240000 This wasn't just an incremental improvement. 214.240000 217.680000 It was a breakthrough moment that shocked the AI community. 217.680000 223.560000 For the first time, deep learning demonstrated that with enough compute power and data, neural 223.560000 229.040000 networks could achieve unprecedented accuracy in complex tasks, like image recognition, 229.040000 230.840000 for instance. 230.840000 232.040000 And what made it possible? 232.040000 235.760000 It was Nvidia's GPUs. 235.760000 240.480000 Researchers realized that training neural networks on GPUs was exponentially faster than 240.480000 242.480000 using CPUs. 242.480000 248.720000 It was cutting training times from weeks down to days, or even hours in some cases. 248.720000 254.400000 This speed unlocked the practical viability of deep learning models across industries. 254.400000 258.720000 That moment sparked the AI boom that we now know today. 258.720000 265.560000 Suddenly, startups, big tech companies, and research labs worldwide needed Nvidia GPUs 265.560000 270.880000 to train their models, whether it was for computer vision, to enable face recognition 270.880000 278.080000 and self-driving car perception, or speech recognition for powerful virtual assistance 278.080000 280.840000 late Surrey in Alexa. 280.840000 287.480000 Natural language processing, which underpins models like chat GBT and Gemini, Nvidia became 287.480000 291.440000 the hardware backbone of modern AI. 291.440000 297.440000 In effect, CUDA turned Nvidia GPUs from gaming accessories into essential infrastructure for 297.440000 303.520000 the AI revolution, positioning the company as a silent force powering the next era of 303.520000 305.760000 technological advancement. 305.760000 308.280000 Now here's where Nvidia was very smart. 308.280000 310.360000 They didn't just sell chips. 310.360000 313.480000 They built themselves an ecosystem. 313.480000 320.180000 They invested in developer tools, software libraries, and platforms like CUDA and TensorRT. 320.180000 325.240000 This made it easier for AI researchers and companies to use their hardware without rewriting 325.240000 327.240000 their entire code base. 327.240000 330.560000 Imagine trying to switch from Nvidia to another chip maker. 330.560000 335.360000 You'd have to retrain your models, rewrite your software stock, and potentially lose performance 335.360000 337.160000 benefits. 337.160000 341.720000 This ecosystem lock-in gave Nvidia deep roots. 341.720000 346.200000 Their presence so entrenched that rivals struggle to uproot them. 346.200000 351.520000 They also diversified their data center GPUs, professional visualization for industries 351.520000 357.440000 like automotive and architecture, and high performance computing for supercomputers. 357.440000 361.320000 So now that you have the background, let's fast forward to today. 361.320000 368.120000 With AI exploded in late 2022 with chat GPT's public launch, every company suddenly wanted 368.120000 374.040000 to integrate AI, whether it was for coding assistance or a creative tools, customer service 374.040000 377.920000 bots, or internal data analytics. 377.920000 384.680000 Training large language models like GPT-4 or Gemini requires thousands of GPUs running 384.680000 391.240000 for weeks, but even using the model after it's trained, what we call inference, requires 391.240000 396.360000 massive computing power, especially when millions of people are interacting with it at the same 396.360000 397.880000 time. 397.880000 404.240000 Nvidia's H100 GPUs built on their hopper architecture are currently the gold standard 404.240000 406.440000 for AI workloads. 406.440000 415.520000 Each one costs between 25,000 to 40,000 US dollars, and big tech companies buy them in racks 415.520000 419.160000 of hundreds or even thousands. 419.160000 425.760000 Nvidia's upcoming B100 chip announced that Computex promises even better performance per 425.760000 429.880000 watt, reinforcing Nvidia's leadership. 429.880000 436.560000 It's no exaggeration to say that today's AI revolution truly runs on Nvidia hardware. 436.560000 438.720000 Now let's look at the numbers. 438.720000 445.200000 Back in 2016, Nvidia's annual revenue was roughly $7 billion driven primarily by its dominance 445.200000 451.480000 in gaming graphic cards and early moves into data center GPUs. 451.480000 457.640000 Fast forwards to last year and that revenue skyrocketed to over $60 billion, nearly a 457.640000 461.120000 tenfold increase in less than a decade. 461.120000 466.840000 That explosive growth has been fueled by the AI boom, with demand for Nvidia's GPUs surging 466.840000 472.480000 as companies race to build and deploy large language models, generative AI tools, and 472.480000 475.720000 advanced machine learning applications. 475.720000 478.860000 But revenue is only a small part of the story. 478.860000 484.520000 Nvidia's market capitalization, its total value on the stock market, has more than doubled 484.520000 490.160000 in the past year alone, briefly surpassing tech giants like Microsoft and Apple to become 490.160000 493.680000 the most valuable company in the world. 493.680000 499.160000 While Apple has since reclaimed this top spot, Nvidia remains firmly planted among the 499.160000 502.440000 top three most valuable companies globally. 502.440000 508.160000 It's remarkable for a company that just a decade ago was seen mainly as a graphics card manufacturer 508.160000 511.120000 for gamers and creative professionals. 511.120000 516.640000 Today, investors view Nvidia as a critical enabler of the AI revolution. 516.640000 522.680000 The role it plays is similar to how Intel powered the PC era with its processors, or how 522.680000 526.320000 Microsoft dominated computing with Windows. 526.320000 532.600000 Nvidia's GPUs are now the essential engines, driving AI training and deployment at every 532.600000 538.040000 major tech company, every cloud provider, and AI startup. 538.040000 543.120000 In other words, if AI is the gold rush of our time, Nvidia is selling the picks in the 543.120000 546.000000 shovels and businesses booming. 546.000000 549.960000 So with all that said, let's address the pessimist in the room. 549.960000 554.480000 Is Nvidia's explosive growth truly sustainable? 554.480000 559.120000 On one hand, AI demand shows no sign of slowing down. 559.120000 562.440000 Enterprises are racing to build private AI models. 562.440000 567.720000 Cloud providers are scaling their infrastructure to support generative AI workloads. 567.720000 573.120000 And new AI startups are emerging almost daily with innovative applications. 573.120000 576.880000 But on the other hand, competition is intensifying. 576.880000 582.760000 AMD, they're developing AI-focused GPUs with highly competitive price-to-performance 582.760000 588.200000 ratios, aiming to chip away at Nvidia's market's dominance. 588.200000 594.440000 Intel, they're investing heavily in AI accelerators and its gaudy chips to regain relevance 594.440000 596.960000 in a data center market. 596.960000 604.400000 Google has its proprietary TPUs or tensor processing units, optimized specifically for its internal 604.400000 607.840000 AI workloads and services like Google Cloud. 607.840000 613.200000 And finally, Microsoft and Amazon, they're partnering with custom chip makers to build 613.200000 618.320000 their own AI accelerators tailored for their massive cloud platforms. 618.320000 626.440000 Meanwhile, AI hardware startups like Cerebrus, Graphcore, and Samba Nova are designing purpose-built 626.440000 632.700000 AI chips that can outperform traditional GPUs in certain specialized workloads, posing 632.700000 636.080000 additional threats to Nvidia's dominance. 636.080000 639.280000 Another risk factor is supply chain dependency. 639.280000 646.120000 Nvidia relies heavily on TSMC and Taiwan to manufacture its most advanced chips. 646.120000 651.760000 Any geopolitical instability or supply chain disruption in that region could severely impact 651.760000 654.800000 Nvidia's ability to deliver products to the market. 654.800000 659.000000 Finally, there's the question of AI monetization. 659.000000 662.560000 Yes, AI is definitely here to stay. 662.560000 663.960000 That's not in doubt. 663.960000 668.960000 And if the AI hype cycle cools down, or if enterprises struggle to generate immediate 668.960000 675.000000 returns from their massive AI infrastructure investments, budgets could tighten. 675.000000 680.680000 That would lead to a slowdown in spending on UGPUs and accelerators. 680.680000 686.280000 So while the long-term outlook for AI remains strong, the real question is, can companies 686.280000 692.040000 monetize AI quickly enough to sustain their current levels of infrastructure spend? 692.040000 698.160000 Because at the end of the day, Nvidia's future growth depends not just on AI adoption, 698.160000 700.680000 but on AI profitability. 700.680000 704.880000 Beyond Nvidia's surging stock price, there's a far bigger trend unfolding. 704.880000 711.280000 AI has sparked a new infrastructure arms race reminiscent of the cloud wars of the 2010s. 711.280000 716.200000 Back then, tech giants like Amazon, Microsoft, and Google, they invested billions to build 716.200000 721.640000 sprawling global data centers, each vying to capture dominant market share in the 721.640000 723.480000 cloud services space. 723.480000 730.880000 Today, that same race is playing out, but this time, it's about AI compute capacity. 730.880000 736.320000 Every major cloud provider is scrambling to deploy massive clusters of GPUs and AI 736.320000 742.720000 accelerators to power the next generation of large language models, generative AI tools, 742.720000 745.440000 and enterprise AI solutions. 745.440000 750.840000 Companies are pouring capital into AI infrastructure, not just to keep up with competitors, but to 750.840000 755.080000 enable entirely new product lines and revenue streams. 755.080000 760.720000 The wave of investment directly benefits Nvidia in the short to medium term, as its GPUs 760.720000 765.040000 remain the gold standard for AI training and inference. 765.040000 769.080000 But it's also triggering intense efforts to find alternatives. 769.080000 774.760000 Open source AI models are being developed to be more compute efficient, reducing reliance 774.760000 776.720000 on expensive hardware. 776.720000 783.200000 New chip architectures are emerging, including neuromorphic computing, which mimics the structure 783.200000 788.200000 of the human brain for energy efficient AI processing. 788.200000 794.040000 And phytonic space chips, which use light instead of electricity to achieve blazing fast 794.040000 797.960000 compute speeds with lower power consumption. 797.960000 803.240000 Startups and research labs are exploring quantum computing applications for AI, aiming 803.240000 809.400000 to solve complex optimization problems beyond the reach of classical GPUs. 809.400000 814.600000 Cloud providers, like Amazon and Google, are also designing their own AI chips, such 814.600000 822.000000 as AWS Trainium and Google TPU, to reduce their dependency on Nvidia and optimize performance 822.000000 824.800000 for their proprietary models. 824.800000 830.800000 In other words, today, Nvidia is king, no doubt, holding the crown as the world's AI hardware 830.800000 832.720000 backbone. 832.720000 837.680000 But the pace of innovation in AI hardware is accelerating rapidly. 837.680000 842.280000 The companies that once competed to build the biggest data centers are now competing to 842.280000 848.960000 build the fastest, most efficient and cost effective AI compute infrastructure. 848.960000 852.760000 And in this race, dominance is never permanent. 852.760000 859.160000 Just as Intel's dominance in CPUs was challenged by ARM and Apple Silicon, Nvidia's leadership 859.160000 862.360000 will face fierce challenges in the years ahead. 862.360000 867.280000 The bottom line for me is this, Nvidia didn't just ride the AI wave. 867.280000 872.680000 They built the surfboard, they trained the surfers, and frankly, they owned the beach. 872.680000 877.200000 They saw where the world was headed and positioned themselves perfectly to lead it. 877.200000 882.680000 Their rise from a niche graphics card manufacturer to become the world's most valuable semiconductor 882.680000 884.960000 company, it's not just luck. 884.960000 893.320000 This is a masterclass in strategic focus, ecosystem building, and investing in the future before 893.320000 895.880000 anyone else saw it coming. 895.880000 900.880000 For tech professionals like us, there are powerful lessons here. 900.880000 904.000000 So first and foremost, stay ahead of trends. 904.000000 907.080000 Nvidia didn't wait for AI to become mainstream. 907.080000 913.560000 They invested in CUDA and general purpose GPU computing years before deep learning exploded. 913.560000 917.440000 You'll ask yourself, where is the next wave coming from? 917.440000 923.440000 Quantum computing, neuromorphic chips, edge AI, synthetic biology. 923.440000 927.280000 Don't just observe the trends, position yourself to ride them early. 927.280000 930.800000 Secondly, build ecosystems, not just products. 930.800000 938.720000 Nvidia's real moat isn't just their GPUs, it's their software stack, the developer tools, 938.720000 943.920000 queries, and partnerships that make their hardware, frankly, indispensable. 943.920000 948.360000 So ask yourself, are you creating standalone products, or are you building ecosystems 948.360000 953.760000 that integrate deeply into your customers' workflows and make them rely on you? 953.760000 956.960000 Lastly, prepare for disruption. 956.960000 961.400000 Nvidia's dominance is impressive, but it's never guaranteed. 961.400000 968.120000 Competitors like AMD, Intel, Google, and AI hardware startups are innovating rapidly. 968.120000 972.440000 So the question for yourself is, what are you doing today to ensure you don't get disrupted 972.440000 974.040000 tomorrow? 974.040000 977.840000 Complacency is the silent killer of market leaders. 977.840000 985.000000 At the end of the day, Nvidia's journey is a reminder that vision without execution is hallucination. 985.000000 989.040000 An execution without vision is directionless. 989.040000 994.840000 It's the combination of both, bold vision and relentless execution that creates industry 994.840000 997.120000 defining companies. 997.120000 998.120000 So what do you think? 998.120000 1002.920000 Is Nvidia's $4 trillion valuation justified? 1002.920000 1006.040000 Or are we heading into an AI hardware bubble? 1006.040000 1008.360000 I'd love to hear your thoughts. 1008.360000 1014.880000 Feel free, please email me at Inspire@techitfromme.com or leave a comment wherever you're listening. 1014.880000 1021.080000 And if you enjoyed this episode, please follow Tech It From Me on Spotify, Amazon, Apple Podcast, 1021.080000 1023.360000 or wherever you get your podcasts. 1023.360000 1025.840000 It really helps the show grow. 1025.840000 1027.360000 This has been Tech It From Me. 1027.360000 1029.880000 Until next time, I'm Mike Madole. 1029.880000 1033.040000 Tech It From Me is an independent and solo-produced podcast. 1033.040000 1034.318313 [BLANK_AUDIO]