Prof G Markets · · 1h 13m

Nvidia’s $2 Trillion Scandal Is About To Get Much Bigger

Summary

Scott Galloway and Ed Elson dissect the NVIDIA scandal, centred on alleged chip diversion to China through intermediaries such as Megaspeed, and what it means for the stock and US policy. Both expect NVIDIA shares to be higher by year-end despite the risks: Galloway argues the company can survive Jensen Huang's exit and that the stock trades cheaply for its growth, while Elson notes a forward P/E near 10 and says he bought around $165. Elson is more alarmed about revenue from diverted sales, estimated by short seller Culper at about 20%, and about the weak board and succession planning, while Galloway calls the board captive and expects Congress and individuals, rather than the company, to bear the consequences. Both see Commerce Secretary Howard Lutnick as exposed to the fallout given his NVIDIA holdings and his evasive answers to Senator Warren. The hosts then turn to Anthropic's planned IPO, which Galloway expects to pop at least 20% on a demand-signal rationale, while Elson warns of heavy compute spending, dependence on fundraising and a small float that could distort early pricing.

Key takeaways

  • Scott Galloway (bullish on NVIDIA): NVIDIA stock will be up by year-end despite the Huang scandal; AI won't collapse without him.
  • Ed Elson (bearish on NVIDIA): NVIDIA's China-diverted chip revenue is likely far larger than Megaspeed; a major risk to the stock.
  • Scott Galloway (bullish on NVIDIA): NVIDIA can survive Huang's exit; it has dozens of capable executives who can lead it.
  • Ed Elson (bullish on NVIDIA): Nvidia is crazy cheap after the scandal selloff; he bought near $165 and sees more upside than downside.
  • Ed Elson (bullish on Anthropic): Anthropic has premium products, remarkable adoption and about 1200% revenue growth; he'd bet on it over rivals.

What was said, by company and theme

24 views from 2 speakers. Every quote is checked against the episode's transcript (how).

NVIDIA

Across the shows: 147 speakers in other episodes, 131 bullish and 110 bearish calls, past 30 days

Where everyone stands on NVIDIA

  • Scott Galloway · NYU Stern School of BusinessBullish

    NVIDIA stock will be up by year-end despite the Huang scandal; AI won't collapse without him.

    Galloway predicted NVIDIA stock would be higher by year-end. He argues the notion that AI collapses without Jensen Huang, or that NVIDIA stops being an amazing company, is wrong. He concedes the broader market might crash AI, so the call is about NVIDIA's own stock rather than the sector.

    “And the second prediction is I think NVIDIA stock is gonna be up. The notion that AI is gonna collapse without Jensen Huang or that NVIDIA isn't... Is not gonna go on to be an amazing company.”at 10:16
  • Ed Elson · Prof G MediaBullish

    Jensen Huang's importance to NVIDIA is overrated; NVIDIA will survive without him.

    Elson says analysts claim NVIDIA would lose 20 to 30 percent of its value without Huang, but he thinks people overrate his importance. He agrees with Galloway's prediction that the stock will be up by year-end and says the company was built around Huang with no succession plan, which is a governance problem rather than proof the business depends on him.

    “I think we're overrating how important Jensen Huang really is, to this company.”at 12:06
  • Scott Galloway · NYU Stern School of BusinessBearish

    NVIDIA's board is not independent and lacks national security expertise, so governance oversight is weak.

    Galloway argues the board is not a strong, independent governance body. Members have average tenure of nearly 15 years, none are former S&P 500 CEOs, the lead director is a 77-year-old former law firm executive, and none have national security experience. Most members have made far more money from Huang than from the rest of their careers combined, so they were effectively in his pocket. He argues the independent counsel review would be paid for by the company and so would not be credible.

    “Almost every person on this board has made more money because of Jensen Huang than they have the rest of their career combined.”at 15:20
  • Ed Elson · Prof G MediaBearish

    NVIDIA's China-diverted chip revenue is likely far larger than Megaspeed; a major risk to the stock.

    Elson cites research from the short seller Culper estimating that about 20% of NVIDIA's revenue comes from diverted China sales. He points out that Singapore, Malaysia and Thailand, the number two, four and five importers of AI chips, imported about $80 billion of chips this year. He argues Megaspeed is only visible because its CEO had an affair with Huang, so the problem likely spreads much wider. He calls it the largest risk for the stock, because it may call into question NVIDIA's revenue from outside the US and allied partners.

    “there's no reason in my view to believe that... To not believe that this is the tip of the iceberg and that this actually spreads a lot wider than we think.”at 20:49
  • Scott Galloway · NYU Stern School of BusinessBullish

    NVIDIA can survive Huang's exit; it has dozens of capable executives who can lead it.

    Galloway argues there are about 40 potential CEOs already inside NVIDIA and that the 'singular Jesus-like innovator' idea is wrong, citing Apple after Steve Jobs, Microsoft after Bill Gates and Alphabet after its founders, all of which did fine. He adds that if NVIDIA can ring-fence the scandal and show real progress with the security apparatus, the company can recover once Huang is out.

    “There are 40 potential CEOs already in the building at NVIDIA.”at 36:03
  • Scott Galloway · NYU Stern School of BusinessBullish

    Nvidia will not fail without Jensen Huang; no single leader is indispensable, and the CFO-as-interim-CEO move was smart.

    Galloway argues the idea that Nvidia cannot survive without Jensen Huang is ridiculous. He says no company is so dependent on one person, and that large companies are run by CFOs, CMOs, supply chain and strategy heads who manage big divisions. He says a board putting the CFO in charge as interim CEO signals 'don't panic' and is a smart move, because the CFO is the source of truth on numbers while CEOs tend to embellish.

    “There is no singular Jesus in business... It's just the notion that the sector in NVIDIA is not gonna survive without this guy is ridiculous.”at 37:39
  • Ed Elson · Prof G MediaBullish

    Nvidia can thrive without Jensen Huang; mature companies run on management, operations and moats, not founders.

    Elson argues markets overvalue the face of a company. He cites Apple after Steve Jobs (the stock roughly tripled relative to the S&P and gained about $5 trillion in value under Tim Cook), Google after Page and Brin stepped back (+$2.7 trillion), and Microsoft after Bill Gates left (more than double the S&P's return). Visionary founders matter for storytelling early on, but a mature company with about $4 trillion in value, the largest market share in the hottest industry, secure supply chains and moats depends more on management and operations, so Nvidia should keep running after Huang.

    “the less it becomes about having a visionary storyteller, and the more it becomes about having great management, having good operations, having secure supply chains, moats”at 40:13
  • Ed Elson · Prof G MediaBullish

    Nvidia is crazy cheap after the scandal selloff; he bought near $165 and sees more upside than downside.

    Elson says he bought some Nvidia around $165 after the scandal-driven drop. Forward P/E has fallen to about 10, below Procter & Gamble, Walmart, Ford and Coca-Cola, which he calls crazy cheap. He acknowledges huge risks and that the stock could fall further, but says the upside risk is much greater. He expects that in about three quarters the earnings will look fine and investors will focus on fundamentals again, re-rating the stock.

    “The fact that the forward PE has come down to around 10, it's trading at a lower multiple than Procter and Gamble and Walmart and Ford and Coca Cola. Like, this is crazy, crazy cheap right now.”at 41:57
  • Scott Galloway · NYU Stern School of BusinessBullish

    Nvidia's multiple is too low for growth above 100%; it trades like a 2% grower automaker.

    Galloway compares Nvidia to companies growing around 4%. Nvidia is growing more than 100% year over year, yet trades at the same multiple as an automobile company growing 2% a year or flat. He acknowledges autos are cyclical, but argues the gap between growth and multiple makes the stock look undervalued.

    “Growing more than a 100% year over year. Okay. 100% year over year. And it's trading at the same multiple as an automobile company growing 2% a year or maybe negative or flat.”at 42:26
  • Scott Galloway · NYU Stern School of BusinessBullish

    Government won't kneecap Nvidia over the scandal; individuals may be charged, but the company survives.

    Galloway argues that Republicans and many Democrats, despite anger, won't want to nationalize or overregulate the company. They'll go after individuals, possibly force the board to resign, or bring charges against Jensen Huang, but won't hurt a company that has created so much shareholder value and is one of the premier companies in the world. People will pay the price, but the company won't be damaged.

    “they're not gonna wanna kneecap our thoroughbreds... This company has created so much shareholder value, is kind of arguably one of the premier American companies”at 43:24

Anthropic

Across the shows: 114 speakers in other episodes, 166 bullish and 85 bearish calls, past 30 days

Where everyone stands on Anthropic

  • Scott Galloway · NYU Stern School of BusinessBullish

    Anthropic's IPO matters more to AI than the Nvidia scandal; a strong listing signals continued chip demand.

    Galloway argues the Anthropic IPO has more impact on the AI sector than the Nvidia scandal. If Anthropic goes public, can access cheap capital, and keeps growing, it will signal that the market still wants more of Nvidia's chips. That demand signal is what matters for AI, not the scandal details.

    “If Anthropic goes public and can access cheap capital and their business keeps growing... creates a signal that we're still gonna want more of those delicious NVIDIA chips”at 50:50
  • Scott Galloway · NYU Stern School of BusinessBullish

    Anthropic's IPO will pop 20-50% on the first trade, because bankers engineer pops.

    Galloway says bankers use AI to map demand in the order book and price the deal so it gets at least a 20% pop. Anthropic's board accepts dilution and raising less money in exchange for a headline-positive debut, so the IPO will be priced to pop out of the gate. He calls the IPO pop the greatest risk-free trade on the upside and says he is trying to get shares, though he has not got them yet.

    “I think it's up 20 to 50% on the first trade because this is what the bankers do.”at 55:58
  • Ed Elson · Prof G MediaBearish

    Anthropic's small float and engineered pop make it underwhelming below $2 trillion; true price comes later.

    Elson says Anthropic will issue a small float of about 6% versus the 15-20% average. That allows financial engineering and avoids true price discovery, so the stock will be pushed up on day one and week one, with the real price taking longer to reveal. He will measure success against the original $2 trillion target. Pricing around $1.4 trillion that rises to $1.8 trillion would not be a fantastic offering, since it sits below what the company wanted even after a pop.

    “they're gonna issue a very, very small float. It's gonna be about 6%. The average IPO is around 15 to 20%.”at 58:29
  • Ed Elson · Prof G MediaBullish

    Anthropic has premium products, remarkable adoption and about 1200% revenue growth; he'd bet on it over rivals.

    Elson's bull case: Anthropic has built the most premium products in the hottest technology, with remarkable adoption and roughly 1200% year-over-year revenue growth. He takes the $81 billion revenue run rate with a grain of salt but still sees strong growth. He says that if you were betting on one company, it would be Anthropic, and that OpenAI loses roughly twice as much money, with an operating loss of more than $21 billion last year.

    “The adoption rate has been quite remarkable as well. The growth rate has been, frankly, ridiculous.”at 1:00:40
  • Ed Elson · Prof G MediaBearish

    Anthropic's compute costs will soar and it depends on fundraising; profitability is far away.

    Elson's bear case: Anthropic plans about $900 billion of compute spend over several years, with roughly $500 billion due by end of 2030, meaning more than $150 billion next year versus about $25 billion this year. Costs grow about 500% while revenue grows around 100%. Compute comes from competitors, forcing it to build its own expensive data centers. It is overly dependent on fundraising and not close to profitability, so a downturn would leave it in trouble.

    “my concern for this company is that they are overly dependent on fundraising. They have to go to the capital markets in order to continue and achieve what they wanna achieve”at 1:03:20
  • Scott Galloway · NYU Stern School of BusinessBullish

    Anthropic at about 19x revenue is a good buy if its growth keeps doubling, and it can fix profitability later.

    Galloway argues growth is what matters. At about $1.5 trillion valuation and an $81 billion revenue run rate, Anthropic is about 19x revenue. If it doubles for two more years, a buyer today would effectively be paying about 4x revenue, lower than Abercrombie & Fitch. He compares this to SpaceX going out near 100x revenue for a company growing in the teens. He's convinced that if Anthropic maintains its growth rate, it can figure out profitability.

    “that means if it doubles for another two years, that means you're having a chance to buy this company at four times revenue.”at 1:04:32
  • Ed Elson · Prof G MediaBullish

    At Nvidia's multiple, Anthropic would be worth about $1 trillion, a good entry point despite the risks.

    Elson used Nvidia as a comparable. If Anthropic traded at Nvidia's multiple, it would be about $1 trillion in market cap, which he sees as a good entry point. He still says the company could lose everything because it has not found a sustainable business model, and that the entry price is what makes the investment worthwhile.

    “if Anthropic were trading the same multiple as Nvidia... it would be around a trillion dollars in market cap. And to me, that's like a good entry point for this company.”at 1:05:40

US–China AI & chip race

Where everyone stands on US–China AI & chip race

  • Ed Elson · Prof G Media

    US AI chip export controls are incoherent, flip-flopping with Trump, which encouraged NVIDIA's complacency.

    Elson argues export restrictions on GPUs have gone up and down because Trump cannot make up his mind. He says Trump's policy seems to depend on who calls him, so Huang came to believe export policy was stupid, that China was okay, and that diversion did not matter. He sees this as the source of NVIDIA's complacency.

    “it's gone up and down and up and down because Trump can't make up his mind”at 20:12
  • Ed Elson · Prof G Media

    Commerce Secretary Lutnick repeatedly blocked action against NVIDIA over chip diversion; he's in hot water.

    Elson says Trump approved taking action against NVIDIA several times, but each time Commerce Secretary Howard Lutnick stood in the way. He notes Lutnick's largest holding is NVIDIA, and that Lutnick gave an evasive answer when Senator Warren pressed him on why Megaspeed had not been added to the blacklist. Elson expects Democrats to investigate him and says he is in very, very hot water.

    “But each time, one person was standing in the way, and that person was Howard Lutnick.”at 23:08
  • Scott Galloway · NYU Stern School of Business

    The administration knew of the chip-smuggling risk and let it slide to protect the S&P 500.

    Galloway says the word is that officials knew the Chinese-linked executive was a spy and did not stop the flow of chips for fear it would hurt the S&P 500. He treats the administration's focus on markets as a national security failure, and he expects congressional hearings to make this the biggest part of the story.

    “the word is they even knew... They knew she was a spy, and they didn't stop it for fear it might get in the way of the S and P.”at 26:28
  • Scott Galloway · NYU Stern School of Business

    AI companies are the new uranium, poorly secured; the administration is incompetent at recognizing the threat.

    Galloway argues AI companies are the new uranium or weapons programs, with Department of Defense and Energy level sensitivity, yet they have been lax on security. He says tech companies likely have not implemented the security their information warrants, and that the administration has been incompetent in recognizing the threat. He expects that once a new layer of security and scrutiny arrives, people will find things that were going on unnoticed, like a house with its back door open and valuables missing.

    “we have been totally lax with what is the new uranium, the new weapons programs, and that is these companies... we have an administration that has been incompetent recognizing the threat here.”at 1:09:00
  • Ed Elson · Prof G Media

    Howard Lutnick's involvement in the Nvidia national security scandal looks significant; he'll likely be out within three months.

    Elson says Lutnick's involvement in the Nvidia national security issue appears significant and really bad, and that he's going to be in hot water. He predicts Lutnick will be out in the next three months. The reasoning is based on how his involvement appears in the scandal.

    “My prediction is that Howard Lutnick is out in the next three months. I think his involvement in this NVIDIA national security thing... Appears significant and really bad, and I think he's gonna be in hot water.”at 1:10:55

AI bubble & tech valuations

Where everyone stands on AI bubble & tech valuations

  • Ed Elson · Prof G Media

    The 40-50% AI share of the market is overstated; giants rebranded as AI, explaining market resilience.

    Elson argues that many large companies are not pure-play AI companies. They already had solid core businesses and rebranded as AI companies by investing in AI, unlike the dot-com era, when internet firms were the entire business. This explains why Nvidia could fall more than 15% in a week while the S&P rose 3%, and why the market has been so resilient despite concentration. He still expects a correction and things could get worse before they get better.

    “they're nowhere near the pure play AI company that you would expect and that you think about when we say things like 40 or 50% of the market is AI.”at 47:45
  • Scott Galloway · NYU Stern School of Business

    Investors rotating out of AI into other stocks is healthy, a slow leak rather than a pop.

    Galloway, citing Robert Armstrong, argues investors are rotating out of AI because they see an uncomfortable amount of risk, but they are not moving into bonds or cash. They are diversifying into other stocks. He thinks this is a healthy recalibration: the AI stocks come down in value slowly, a slow leak of air rather than a crash like the railroad or internet bubbles.

    “Investors are rotating out of AI because they see an uncomfortable amount of risk, but they're not putting it in bonds or they're not putting it under their mattress. They're putting it in other stocks.”at 48:41

Across the shows

What other episodes Chorus covered said about the same companies in the 30 days up to this one.

  • NVIDIA147 speakers · 131 bullish, 110 bearish calls
  • Anthropic114 speakers · 166 bullish, 85 bearish calls

Topics

  • AI chip export controls
  • AI company security
  • AI share of market overstated
  • AI stock rotation
  • Anthropic IPO as AI capital signal
  • Anthropic IPO first-day trading
  • Anthropic IPO pricing and float
  • Anthropic compute spend and profitability
  • Anthropic growth and valuation
  • Anthropic product adoption and growth
  • Anthropic valuation
  • CEO succession risk
  • NVIDIA board oversight
  • Nvidia national security scandal fallout
  • Nvidia valuation multiple
  • chip diversion to China
  • chip export enforcement
  • government response to Nvidia scandal
  • post-Huang leadership
  • stock outlook after CEO exit

More from Prof G Markets

All Prof G Markets episodes

Ask your AI about this episode

Chorus gives Claude, ChatGPT and other assistants every view from this episode and every other show Chorus follows. Ask who agrees, who changed their mind, and what else was said about the same companies.

Setup guide

A weekly email with the calls that mattered from the most important voices in investing. Unsubscribe any time. Privacy Policy

Chorus reports what speakers said. It isn't investment advice. How Chorus works · About Chorus