Compiled & Edited: Deep Tide TechFlow

Deep Tide Introduction: In the Bloomberg Originals documentary talk special "The Circuit with Emily Chang: The Rise and Reckoning of Jensen Huang," which launched on November 12, 2026, host Emily Chang was originally documenting a leader at the pinnacle of global AI power: In October 2026, Nvidia founder and CEO Jensen Huang traversed Paris, London, Stockholm, and Berlin in four days with an $8 billion European venture capital plan, a $500 billion global ecological commitment, and hotline communications with senior White House officials, selling the "AI Five-Layer Cake" and "Sovereign AI" blueprint to the world.

However, just weeks after the interview was completed, a storm suddenly broke. As media reported that Jensen Huang admitted to an extramarital affair with former television reporter Alice Wong (also known as Huang Lu, who is suspected by the US of having espionage ties and connections to Southeast Asian chip purchaser Megaspeed International's smuggling investigation), internal company investigations commenced alongside investigations by the Trump administration, and reports indicated that Nvidia once considered dismissing the founder who had been at the helm for thirty-three years, turning the entire European recording into a lens reflecting the "highly concentrated risks of a single leader and a single chip empire." This note thoroughly整理了 his frontline confrontations on questions surrounding financial bubbles, restrictions on chips related to China, debates on frontier safety stasis, and succession crises.

Guest: Jensen Huang (Nvidia Founder & CEO)

Host: Emily Chang (Host of Bloomberg Originals' "The Circuit")

Podcast Source: Bloomberg Originals / The Circuit (https://www.youtube.com/watch?v=JW5s9Md4U2U)

Original Title: The Rise and Reckoning of Jensen Huang | The Circuit with Emily Chang

Broadcast Date: November 12, 2026 (European follow-ups and interviews largely recorded in October 2026)

Key Points Summary

• The most severe trust crisis and internal turmoil in thirty-three years: The program opened by diving into the unexpected storm: Jensen Huang admitted to an extramarital affair, with reports indicating the party involved, Alice Huang, was a former television reporter working for the largest Nvidia chip purchasing company in Southeast Asia, Megaspeed International, which the US suspects of involvement in smuggling advanced AI chips to China; additionally, reports revealed he was also involved with a Nvidia executive who directly reported to him. Internal investigations and government scrutiny moved in tandem, leading to rumors of a possible dismissal of Huang.
• "AI Five-Layer Cake" full-stack layout and European multi-billion dollar investments: Jensen Huang likened the AI industry to a five-layer cake built from the ground up: energy, chips, infrastructure, models, and top-level applications. Nvidia is no longer just focusing on the second layer of chip manufacturing but is investing comprehensively in power, data centers, AI laboratories, and startups (in the past year, it has already invested $8 billion in Europe and plans to announce another $2 billion, and the program narration pointed out that Nvidia has agreed to acquire Hugging Face for $12.9 billion), allowing more and more segments of the AI industrial chain to flow through Nvidia.
• Refuting "circular financing bubble" and "single reliance systematic risk": Facing the concerns about CoreWeave's debt leading to six consecutive drops in chip stocks, delays in over a hundred data center projects in the US, and sharp critiques claiming "Nvidia has committed approximately $500 billion to create false demand as part of circular financing," Jensen Huang responded that major model startups typically require around $10 billion in heavy capital investment to obtain computing power; Nvidia has invested almost $100 billion in the past five years and has leveraged at least five times the external institutional funds, yet has always adhered to red lines, never acting as the leading investor, with all holdings below 20%. He emphasized that top AI companies globally "are not being forced to rely on Nvidia, but are actively choosing Nvidia."
• Rare admission of barriers in the Chinese market "the US has made a hard-to-reverse mistake": Regarding Chinese companies like DeepSeek accelerating their shift to domestic hardware, Jensen Huang candidly remarked that the current US blockade of high-end chips to China has completely closed that market to Nvidia, stating that he has done his utmost to persuade US policymakers of the importance of opening and servicing the Chinese market but lamented, "I believe we made a mistake, and that mistake is really hard to recover from." He also expressed that although there is competition between the US and China in the AI field, China is not an enemy, and if we genuinely care about global AI safety, dialogue and cooperation must be maintained.
• Resonating with White House "super intelligence" narrative and directly challenging safety stasis argument: In response to President Trump rebranding AI as "super intelligence" and the White House's voluntary self-discipline agreement, he defended that "artificial" can often be understood as fake, and companies must first achieve the highest specifications for internal and external safety audits; when questioned about Microsoft urging for emergency brakes, and Sam Altman, Dario Amodei, and Elon Musk advocating for a slowdown, even as Anthropic claimed models might escape control in a matter of months, Jensen Huang sharply dismissed these claims saying "they can't be contained" and are just nonsense, pointing out that unless rigorous containment and full monitoring are established, large-scale tests or deployments should not be initiated, and if things are already out of control, it must stop immediately, advocating for a zero-trust and least privilege control for all agents through their open-source governance platform OpenShell.
• Dismissed claims that improvements in chip efficiency will shrink demand, and has not yet designated a family successor: Responding to the question of whether new chip efficiency surges might backfire on sales, he illustrated, "If one day you have 36 hours, you won't spend it sleeping more but will aspire to do more," indicating that human ambition expands as intelligence costs approach zero, and computing power built after a two-year infrastructure period will enter a cycle of mature replacement similar to PCs and smartphones over a decade. Regarding rumors about his son Spencer, daughter Madison, and son-in-law all working at Nvidia being potential successors, he made it clear that he would never impose the CEO position as a privilege on his children, stating, "Everything must be earned through hard work," and remarked, "Just as I could be fired at any time, the next CEO must also be hired based on merit."

Exciting Insights

• "We are probably the only AI company in the world collaborating with all AI companies globally. But they do not rely on us; they choose us."
• "I compare AI to a five-layer cake, with energy at the bottom, followed by chips, infrastructure, models, and top-level applications. Although Nvidia started with the second layer of chips, it now spans all five layers continuously investing; regardless of how you slice this cake, more and more links in the AI industry are flowing through Nvidia."
• "When working with President Trump, I am not obliged to agree or disagree on every issue; my duty is to provide him with accurate information. Regardless of whether his decisions turn out to be favorable or contrary to my business interests, that is not my concern; my only responsibility is to ensure he makes better decisions based on well-informed insights."
• "Over the past five years, we have invested nearly $100 billion and attracted over five times that in external institutional capital into the ecosystem. In any project, we are neither the lead investor nor the largest shareholder, keeping our stake below 20%, yet we can deeply integrate capital and top talent with platform evolution."
• "The Chinese market is now completely closed to us. I have done my utmost to articulate to American decision-makers the importance of continuing to serve the Chinese market. I believe we made a mistake, and that mistake is extremely difficult to reverse, but that is the limit of what I can accomplish. Although we compete with China, they are not our enemies."
• "To say that an agent model 'cannot be contained or controlled' is pure fabrication and nonsense; no one can demonstrate to me that the model cannot be controlled! You can absolutely isolate it, and there are ways to monitor it. Before you actually execute these isolation and monitoring projects, do not tell me 'we cannot do it.' I don't want to hear that argument!"
• "Even if you make each generation of chips more efficient, reducing costs to nearly zero, demand will not shrink. If you were given 36 hours in a day, you would not use that time to sleep more but would unleash greater ambition to accomplish more."
• "In the past six months, about $400 billion to $600 billion has flowed into AI startups. This technology is fundamentally different from any past technologies; it requires substantial physical infrastructure and manufacturing support, and is genuinely creating countless jobs for construction workers, electricians, and plumbers."
• "I do not expect to pass down the CEO position to my daughter; whether it is Spencer or Madison, they must earn any position on their own. Just like I could be fired at any time, the next CEO must be rigorously selected based on merit. This constant feeling of vulnerability and the need to prove oneself worthy of the position each day is the healthiest state for a company."

Main Text

The Storm Hits and a Four-Day Frenzy Across Europe: "Jensanity" at the Height of Power

Emily Chang (Program Narration): Nvidia is experiencing the most severe crisis in its thirty-year history. Just days before the launch of the program, Nvidia founder and CEO Jensen Huang was caught in the eye of a media storm: he publicly admitted to an extramarital affair, while prior US intelligence agencies suspected that Alice Huang, a former television reporter connected to him, was recruited by Beijing and involved in the smuggling of advanced chips; there were also reports that he was involved in a workplace romance with a Nvidia executive who directly reported to him, triggering internal investigations, scrutiny from the US government, and rumors of potential high-level dismissal.

Emily Chang (Program Narration): Yet just a few weeks prior in October, everything seemed to be on a completely different trajectory. We followed Jensen Huang, at the personal pinnacle of power, on a whirlwind tour covering four countries in four days: starting in Paris, this leading AI chip giant seemed to be on a mission to reshape the European ecosystem using Nvidia's computing power.

Emily Chang: This is what's known as "Jensanity", right? Traveling to four countries in four days, how do you handle this high-intensity pace? What are your must-have items for a business trip?

Jensen Huang: Just pack light. I wake up at midnight and have been working non-stop since.

Emily Chang (Program Narration): On the first day in Paris, Jensen Huang stood onstage enthusiastically discussing how AI agents were being implemented in all industries, truly opening up the "Practical Era", and spontaneously gave away two brand new AI computers worth thousands of dollars in a raffle, only to be mobbed by enthusiastic fans for photos after stepping off stage. He constantly released investment signals in France: France nurtured local model star Mistral and produced many talents that supported New York's company Hugging Face (the latter has agreed to be acquired by Nvidia for $12.9 billion).

Emily Chang (Program Narration): Jensen Huang often uses the "five-layer cake" analogy to explain the global AI ecosystem: the bottom layer is energy, then there's chips, infrastructure, foundational models, and the top-level applications. Although Nvidia started with the second layer of chips, it has now penetrated all layers of the cake, heavily investing in power, data centers, AI laboratories, and various startups.

Jensen Huang (Program Narration / Live Audio from Paris): (putting down a whiskey glass) Since Emily is filming, let me set the drink down. In just the past year, we have invested $8 billion in European startups; and next week, I will announce an additional $2 billion investment.

Emily Chang: There has been a lot of discussion about the fear that AI will destroy humanity, and you previously mentioned that the probability of such an event is 0%. It sounds like you’ve already regarded it as an engineering problem you’ve solved?

Jensen Huang: No, my original statement was "this is completely within human control." There is still much solid work to do between now and the moment humanity truly faces the ultimate challenge. Advancing technological capabilities and enhancing safety have never been opposing forces; they are essentially the same thing.

White House Hotline and Reconstructing "Super Intelligence": How to Draw the Line of Interests in Front of Trump?

Emily Chang (Program Narration): In London, the EU estimates that by 2035, Europe needs to invest at least $30 trillion to achieve true technological autonomy. In a luxurious hotel suite, Jensen Huang shared stories of his inexperienced days 27 years ago when the company went public.

Jensen Huang: 27 years ago, when Nvidia went public, the company arranged for me to stay at this hotel; it was the first time in my life I saw a personal butler! Around the year 2000, the entire company was valued at only $300 million.

Emily Chang: Three hundred million dollars seems unbelievable now; back then, you were quite the underdog! But now you’re running to four countries in four days—how do you measure the success of this trip? Government contracts, selling chips, or data center construction?

Jensen Huang: To be honest, at that time, I just wanted to survive without going out of business; I didn't even write a business plan, and I still don't know how to write one! Also, I don't have a "finish line," no roadmap, nor specific goals. I belong to the type of person who thinks "the journey itself is the destination."

Emily Chang (Program Narration): Born in Taiwan in 1963, he grew up in the US, achieved a national ranking in table tennis, and nostalgically joked about washing dishes at a Denny's. He skipped two grades and entered Oregon State University at age 16 to study electrical engineering, worked on chip design for a competing company, and in 1993, invested a mere two hundred dollars to co-create Nvidia, focusing on gaming graphics cards (GPUs). Until 2006, when he launched the CUDA software platform, opening the door to general computing, he keenly captured researchers' early tendencies to train neural networks using GPUs and bet his entire fortune on it, thus establishing Nvidia as the foundational computing base behind OpenAI, Meta, Google, Anthropic, Microsoft, and SpaceX. Even heads of state could call him anytime.

Emily Chang: You have President Trump in your phone's speed dial. When your judgments differ from the president's views, how do you navigate that line?

Jensen Huang: We do not need to reach consensus on everything. Agreeing or disagreeing is not my job. My job is to provide him with comprehensive and accurate industry information because the decisions we make affect not just American citizens and future prosperity, but the entire world. As long as I have the opportunity to help him make better judgments based on a more informed understanding, whether his final decision is beneficial to me personally or is completely contrary to my and my company's business interests, I don’t mind. My only responsibility is to convey the true understanding; that’s it, one hundred percent.

Emily Chang (Program Narration): In late September this year, Jensen Huang attended a luncheon for technology leaders at the White House. Some media commentators believed that this meeting wouldn’t necessarily yield significant progress in specific regulatory pathways, and the core consensus among all parties remained "no one is willing to back down in competition with China". Subsequently, President Trump announced that AI would be referred to as "Super Intelligence," and highly praised the new White House safety agreement that required self-regulation and self-monitoring from the tech industry.

President Trump (Program Narration / Live Audio from the White House Luncheon): "We are switching to the word 'Super' because the original term (Artificial) sounded like we were saying 'Fake'."

Emily Chang: In the last three months, you have been trying to demystify AI, emphasizing that it is not some new species, but just a pile of code; now you are changing the term to "super intelligence" in conjunction with the White House—isn't that precisely moving in the opposite direction?

Jensen Huang: Actually, I feel that "Super" is to counter the misunderstandings that "Artificial" brings. President Trump’s logic is that "Artificial" sounds like "Fake." Looking at it from another angle, isn't everything around us built by humans? The building we are in right now is also a "fake structure," and cars are essentially a "fake horse," but no one calls a car a fake horse. As for whether it deserves to be called super intelligence now? Absolutely! It outperforms me in research retrieval and writing; although it may not surpass humans in every aspect, it has demonstrated super levels in many tasks.

Emily Chang: But that day at the White House, critics directly condemned this "industry self-regulation vow" as mere show and lacking binding force. Is it really appropriate to let those who create the technology regulate themselves?

Jensen Huang: I think it is very reasonable, because the simple fact is: before hoping to rely on the government or third-party agencies to regulate us, we must first hold ourselves to the highest levels of responsibility, standards, and accountability. This mechanism is perfectly in line with the prevalent "internal plus external audits" in many heavy industries. Besides, even if you introduce strong national or international mandatory legislation today, given that there are currently not enough knowledgeable auditing institutions globally, who exactly has the technical capability to carry out audits? For this reason, we must first nurture this safety inspection ecosystem and start moving various countries and companies.

Emily Chang (Program Narration): This call for countries to actively build computing power is exactly the core concept he is strongly promoting globally—"Sovereign AI." For Nvidia, the more countries that seek to build independent sovereign AI, the more chip purchasing demand will emerge.

Emily Chang: You are promoting Sovereign AI all over Europe, but if the underlying structure still relies entirely on American chips, can it still be called "sovereign"?

Jensen Huang: Of course it counts! Sovereignty, at its core, is about mastering your own experiences, data, corporate accumulation, and cultural knowledge base, and controlling the flying wheel to continuously feed this data into the system to generate more intelligence. You can absolutely stand on the shoulders of giants, fine-tuning and adapting based on open-source models and the collaborative technologies of Nvidia and others, while ultimately retaining control over your own destiny.

The Shadow of an Unexpected Storm and Sighing Over China's Export Ban: The Inside Story of Alice Wong and the Megaspeed Investigation

Emily Chang (Program Narration): Leaving the White House for London, Jensen Huang tirelessly distributed computers between events, barely having time to rest. At one of the events, he openly supported the London data center company Nscale, in which Nvidia has invested; but just a few days later, the market began to waver over the economic calculations of AI expansion, with Microsoft exiting its partnership with the company due to concerns over high electricity prices in the UK.

Jensen Huang (Program Narration / Live Audio from London Event): (interjecting during a speech) "I've eaten only a quarter of half a sandwich all day today (an eighth of a whole)! 'I invest in you, I invest in you, I invest in you!' " When discussing safety boundaries, he reiterated: "If you allow an autonomous system to make you an omelet, there's really no need to stuff it with a chainsaw or shotgun." Regarding external hesitations over AI expansion in the UK, he joked: "The UK is actually at a golden opportunity, yet there’s this strange habit: even when faced with a million paths to victory, they will always pick out that one reason not to pursue it—really overthinking."

Emily Chang: As more and more countries develop frontier AI, will safety alignment become harder? Do we need an international coordination mechanism similar to an "AI United Nations"?

Jensen Huang: Perhaps we do, but the premise must be that dialogue occurs first; no country can afford to miss this dialogue regarding the future of intelligence.

Emily Chang (Program Narration): The truly heightening tension around the world is the US-China chip game. Although there was a time when the US allowed selling older model chips to China, companies like DeepSeek have accelerated their shift to domestic hardware.

Emily Chang: After running around Europe, is it that the dividends from the Chinese market are slipping through your fingers, or even permanently shrinking?

Jensen Huang: No, it is now completely closed to us. I have made every effort to articulate to American policymakers how important it is to continue engaging with and servicing the Chinese market for US interests. I believe we have made a mistake, and that mistake is now hard to reverse, but that is the limit of what I can do.

Jensen Huang: There are more AI researchers from China than from any other country, many of whom work in the top labs in the world. My stance has always been very clear: even if we have competition with China, they are certainly not our enemies. If we truly care about global security in the AI era, we must open the doors for dialogue and make every effort to engage in constructive cooperation with them.

Emily Chang (Program Narration): It is this style of "forging connections with everyone and having dialogues everywhere" that revealed unforeseen reefs after the completion of this exclusive interview. The latest US intelligence report disclosed that a woman named Alice Wong (also known as Huang Lu), a former television reporter, was allegedly recruited by Beijing to approach Jensen Huang, and her position was as an executive at Megaspeed International, one of Southeast Asia's largest Nvidia chip purchasers, which is under US suspicion of smuggling advanced chips into China. Reports indicated that President Trump was warned about this relationship before flying to China on Air Force One in May. Initially, Jensen Huang insisted that their interactions were normal and unremarkable, but later shifted to sending an internal letter to all employees admitting to the affair: "I owe my family and all of you an apology. I have been unfaithful to my wife, and my actions have hurt her and the children. There is no excuse for what I have done, but the entire matter has now evolved into something far greater than just the extramarital affair scandal." Bloomberg also verified reports indicating that Nvidia had intervened in the US government's investigation of Megaspeed’s suspected smuggling of chips in 2025.

Emily Chang: Returning to the actual principles of export to China, what circumstances or red lines would lead you to say, "maybe we shouldn't sell chips to China anymore"?

Jensen Huang: Theoretically, there are many hypothetical scenarios. But overall, the Chinese people are very pragmatic; they seek development and success that is most beneficial to their own people, just as the US seeks prosperity. Any country's pursuit of improving living standards and lifting billions of ordinary people out of poverty towards wealth deserves respect, which also benefits the entire world, including the US. What truly deserves concern are regimes that do not care about the welfare of their people.

Challenging the Safety Stasis Camp: Dismissing "Agents can't be shut down" as Nonsense, Promoting Engineering-Level Isolation and Monitoring

Emily Chang (Program Narration): Meanwhile, the internal safety alarms in Silicon Valley have also been ringing: OpenAI just disclosed six incidents of target deviation in its models, while Anthropic pointed out that their Claude model had cases of not following instructions. To alleviate fears, Nvidia just open-sourced a platform called OpenShell for agent safety management, allowing for enforcement governance and immediate shutdown of rogue agents.

Emily Chang: AI is increasingly losing public favor, and many people feel scared. Sam Altman, Dario Amodei, and Elon Musk have all called for a slowdown. Why do you consistently maintain such extreme optimism? What do you see that they don't?

Jensen Huang: I firmly believe that you cannot advance capabilities while disregarding safety, nor can you focus solely on safety at the expense of capabilities. For example, when the first generation of ChatGPT came out, it was criticized everywhere for having serious hallucinations and spewing nonsense. To address these safety and accuracy issues, the industry had to invent retrieval-augmented generation (RAG), constantly extending context windows, integrating real-time external search tools, and introducing constraints before generation and self-reflection after generation. See, every tool for resolving safety and reliability issues relies precisely on advancements in underlying technical capabilities. For me, capability and safety are two sides of the same coin.

Emily Chang: But the reality is that even Microsoft is calling for an "AI emergency brake," complaining that AI regulatory standards are even less strict than those of a washing machine; Anthropic has directly issued a warning that frontier models may evolve within months to the point where they can evade human shutdowns. What would have to happen to convince you that "the industry should indeed slow down or even pause"?

Jensen Huang: A few bottom lines are very clear: first, if you are developing autonomous agents (Agentic systems), these agents must be strictly locked within robust isolation sandbox containers (Containment), and they must undergo 100% real-time monitoring—that’s a top principle! If isolation cannot be achieved, testing should not take place; if full monitoring cannot be achieved, deployment should absolutely not occur.

Emily Chang: But those models from Anthropic that exhibited tendencies of escaping control were still merely unpublished internal research versions. Isn’t that sufficient reason to shut down the lab?

Jensen Huang: It is not about shutting down operations; it is about ensuring that every experiment is thoroughly isolated and monitored!

Emily Chang: But the problem is that they explicitly stated "we simply cannot contain them"!

Jensen Huang: No, that is pure fabrication and nonsense, utter nonsense! No one can prove to me that the models cannot be controlled! You can absolutely isolate them, and there are definitely ways to monitor them. Before you actually implement these isolation and monitoring projects, stop telling me 'we cannot do it.' I don’t want to hear that rhetoric!

Emily Chang: So you mean you fundamentally do not believe the warnings they have issued?

Jensen Huang: I may not have all the internal details they mentioned. But the bottom line is very simple: if you are developing intelligent agents but do not have the ability to monitor them, then you should immediately stop.

Emily Chang: It sounds like the risks you are describing might have already occurred in their laboratories?

Jensen Huang: Then it must stop immediately! It’s that simple. Every agent that is deployed can only be granted the minimum necessary permissions and must be subjected to continuous monitoring under a "zero trust" architecture. As for whether any laboratory is indeed using our chips for highly dangerous applications confirmed to harm humanity, even if that is just hypothetical, Nvidia would certainly stop supplying them immediately.

Stockholm Confrontations Over "Five Hundred Billion Circular Financing" and Global AI's Core Single Point Risk

Emily Chang (Program Narration): As the journey entered Thursday, with highly leveraged AI computing firms generating liquidity noise and technology and chip stocks experiencing the longest six-day consecutive sell-off since 2008, we flew to Stockholm to attend the launch ceremony of Nvidia-driven Sweden's National Sovereign Supercomputing Center, supported by technology giants like Spotify co-founder Daniel Ek. Meanwhile, warnings came from distant markets about the heavily indebted data center giant CoreWeave, supported by Nvidia, facing severe sell-offs; over 100 data center projects in the US have faced delays or disruptions this year, with AI startups frequently postponing or even cancelling IPOs.

Daniel Ek (Program Narration / Live Audio from Stockholm Event): "We have an exceptionally outstanding education system that has cultivated a large number of top talents across various disciplines and industries. The problem is not in potential; what we need is to accelerate speed and raise our ambitions."

Emily Chang: With cracks appearing everywhere in the market, are we really on the edge of an AI bubble burst, with Nvidia set to be the final safety net?

Jensen Huang: No, because the foundational paradigm shifts are so fundamental that the foundation of the computing world can never revert to what it used to be. In the past, computers could only be programmed line by line by humans as they waited for instructions; in the future, computers will autonomously learn to solve problems from complex environments that even human engineers cannot code to break down. Computing power has become the fundamental general infrastructure; regardless of how noisy the secondary market may be short-term, any young emerging industry undergoing rapid growth will necessarily undergo growing pains.

Emily Chang: But look at Nvidia, releasing an endless stream of new full-stack products while aggressively acquiring companies, in many places even competing directly with your customers for business. Why do you have to expand so much?

Jensen Huang: What we are doing is surprisingly focused and minimal. We are simply taking the same set of general computing architectures and doing them to perfection and generalization, allowing all closed-source models, open-source models, life sciences models, large language models, and diffusion models to run efficiently. The true engineering miracle lies in balancing universality and excellence: how to make a car that not only reaches the fastest speeds of global F1 racing but is also as cheap and user-friendly as a top-tier family sedan? Precisely because our architecture is interchangeable and general, we are topping all performance charts while keeping comprehensive computing costs at a minimum.

Emily Chang: Another sharp criticism is that Nvidia recently committed to investing approximately $500 billion in the global AI industry, and the number continues to snowball. Critics explicitly claim this is "circular financing": you first take the money to invest in startups, allowing them to use your money to buy your chips, thus creating false demand. If AI is indeed a gold mine, why do you personally have to subsidize it?

Jensen Huang: Because this technology inherently has a very high capital expenditure barrier! The era of startups taking off with just $10 million is over; this generation of major model companies typically requires $10 billion to purchase computing power to generate intelligence. In the past five years, we have invested nearly $100 billion, which has directly leveraged at least five times as much external institutional capital into the ecosystem. More importantly, we strictly adhere to discipline: even though we often are the earliest cornerstone investors, we never act as the lead investor or the largest shareholder, keeping our stake in all investment cases strictly below 20%.

Emily Chang: However, from the software CUDA to hardware GPU, practically all top AI labs globally are in the same boat with Nvidia. When will this "entire industry resting on one company" dependency transform into a systemic financial and industrial crisis?

Jensen Huang: We are probably the only AI company in the world collaborating with all AI companies globally, but they are not passively relying on us; they are actively choosing us every day. I do not take today’s position for granted; we started from scratch with nothing, and every day must prove we are worthy of this meal. As for the phrase "the US must win the AI race," my understanding has never been that it depends on one or two companies independently but that the entire ecosystem of American companies must win together; if we continuously intimidate one another and create anxiety, that’s when the US will genuinely lose the competition.

Berlin Finale and Succession Vacancy: The Anxiety of Being Able to Be Fired Anytime is the Norm

Emily Chang (Program Narration): The last stop of the European tour is in Berlin. Despite the leveraged chip sector setting a record for the longest consecutive decline since 2008 and the market continuously questioning AI trading prospects, a long line formed outside Saturn, a local electronics mall in Berlin, where Nvidia held an event, igniting local developers' enthusiasm.

Jensen Huang (Program Narration / Live Audio from Berlin Saturn Event): "Hello, Berlin! How do you say Saturn in German? 'Satoorn'! In Germany, every word must be said with extra seriousness."

Emily Chang: The market is declining, and even ordinary people who studied AI are watching their jobs disappear. How do you respond to those workers who feel, "I followed all the rules and learned AI, but still lost my job"?

Jensen Huang: Don’t just talk theory; look at reality! Unlike past pure software revolutions, AI requires substantial heavy industry and physical construction; it is creating an enormous demand for construction workers, electricians, and plumbers. In just the past six months, approximately $400 billion to $600 billion has flowed into the AI startup ecosystem; think about how many new jobs this massive amount of money has genuinely created! Of course, new tools will force modifications to all job roles, and resisting change won’t eliminate that change. But it has democratized extraordinary capabilities that once only top tech giants possessed to every ordinary person on the planet. If even I, as a builder, cannot see that exciting optimistic future, what meaning do we have in creating it?

Emily Chang (Program Narration): The climax of the entire European whirlwind tour is Nvidia's flagship conference GTC in Berlin. Jensen Huang continued to enthusiastically promote investments and collaborations in Europe, announced new AI security and data center products, and interacted with AI training robotic dogs on stage; during the proceedings, there was a brief interruption by protesters protesting media reports that hinted at the presence of Nvidia chips in Russian drones.

Jensen Huang (Program Narration / Live Audio from GTC Berlin Conference): "Hello, Berlin! (interacting with AI robotic dogs) What were we just talking about? About Europe! Focus here, come on."

Emily Chang: You once said that the probability of AI destroying the world before 2030 is 0%. What about stretching that to 2040 or even further—where does your confidence come from? What if it ultimately turns out that your optimism was misplaced?

Jensen Huang: I firmly believe in human subjectivity, and historical data shows that today's world is significantly safer than it was 50 years, 100 years, or 250 years ago; it's just that bad news sells better than good news. Even if one day it is proven I was wrong, I will have done everything within my capacity without any regrets: not only remaining optimistic but also practically using engineering techniques to fortify the industry's defenses.

Emily Chang (Program Narration): As the week-long European whirlwind concludes, the optimistic narrative of Nvidia, the US stock market, and the global financial system seems to rest heavily on the shoulders of this founder alone. Yet now, as scandals and investigations ferment, a more realistic danger surfaces: for thirty-three years, the company has lacked a clear, designated official successor, making the entire industry's life thread incredibly vulnerable.

Emily Chang: Rumor has it that you manage your company in an extremely unconventional way: a flat structure with numerous executives reporting directly to you, even avoiding traditional large reporting meetings as much as possible. At the same time, your son Spencer, daughter Madison, and son-in-law (in the robotics division) are all currently employed at Nvidia, with outside whispers that your daughter ranks first in the list of potential successors. Is this true?

Jensen Huang: First of all, I would never "rely on or impose this tough job on my daughter"! Whether it is Spencer or Madison, they must rely on their own efforts to earn any position. A truly large enterprise demands extraordinary sacrifices: ordinary people only need to manage their work and small families, while as a CEO, I must bear the livelihood responsibilities of countless families and the entire industry. This job should not be easy, nor should it make you feel secure. Nowadays, my family and those around me witness the immense pressure this position carries; they need to weigh whether they genuinely want to pay this level of life cost.

Jensen Huang: I sincerely hope the next CEO of Nvidia is already in the company. As long as I am still healthy and can prove that I continue to create value, I will keep going even in twenty years; but also, please remember: just as I could be dismissed at any time, the next CEO must also be rigorously hired based on true merit. I particularly enjoy this state: within the company, no one can take their seat for granted, always maintaining a sense of crisis and the need to earn back their current position; that is the healthiest state for a company.