
Nov 13, 2026 · 1h 7m
NVIDIA leadership risk collides with an AI valuation reckoning
Jensen's Shocking Ouster (What Now?); Neocloud Madness; $RKLB, $ASTS, $ABNB, and $MELI Earnings
The episode connects questions about Jensen Huang’s leadership and NVIDIA’s geopolitical exposure with a broader test of AI infrastructure economics and investor discipline.
- 1NVIDIA’s reported leadership controversy raises questions about board accountability, succession risk, and the durability of its AI-chip advantage.
- 2Airbnb and MercadoLibre show contrasting paths to durable growth, while Snap exposes how stock compensation can undermine apparent profitability.
- 3Neocloud losses, space-industry capital needs, and possible GPU-cycle fatigue challenge the assumption that AI and frontier technology growth will remain frictionless.
Don't miss
The hosts connect the reported Jensen Huang controversy to a broader question: could leadership disruption and geopolitical pressure expose risks in NVIDIA’s AI-cycle dominance?
The brief
Ryan Henderson and Brett open with the reported controversy surrounding Jensen Huang, asking what the board knew, whether shareholders deserved a vote, and how leadership risk could affect NVIDIA.
The discussion widens into geopolitics: alleged Chinese connections, chip-export pressure, Taiwan and TSMC, and the possibility that AMD, hyperscalers, or alternative providers could benefit.
Earnings analysis shifts to Airbnb and MercadoLibre, where growth, margins, competitive advantages, valuation, currency, and AI-agent disruption matter more than headline momentum alone.
Snap’s limited GAAP profitability becomes a case study in dilution, while CoreWeave and Nebius face scrutiny over losses, debt, capital spending, and customer prepayments.
The closing stretch tests the AI and space cycles: Anthropic’s potential IPO, Micron versus NVIDIA, Remitly’s operating leverage, and Rocket Lab’s growth against financing and launch constraints.
What was said on this episode
41 statements · 15 positive · 24 negative · 2 neutral
Megaspeed aimed to facilitate chips and data centers for Chinese customers in Southeast Asia.
“The goal, I believe, was becoming a neocloud or essentially being a facilitator to get chips and data centers in Southeast Asia for Chinese customers such as Alibaba.”
Listen at 3:55
Megaspeed was viewed as a potential channel for China to bypass sanctions.
“This put it on the radar of the U.S. State Department as a potential way for China to bypass sanctions.”
Listen at 4:07
NVIDIA security services paid a bystander who witnessed Huang and an executive.
“NVIDIA security services paid off the bystander.”
Listen at 5:29
Jensen Huang was forced to resign as NVIDIA CEO.
“Jensen was forced to resign.”
Listen at 7:14
Additional cases involving Jensen Huang will probably emerge.
“My guess is where there's smoke, there's fire, and that there's probably going to be a couple of cases, maybe more, that come out, frankly.”
Listen at 8:37
NVIDIA’s board likely has more information and will ultimately be vindicated.
“I am guessing that they are privy to more information than everyone else and that they will probably come out on the right side of history on this one.”
Listen at 10:15
Jensen Huang’s departure could materially harm NVIDIA.
“There is a world where this results very badly for NVIDIA.”
Listen at 11:50
The controversy will have a negative outcome for NVIDIA.
“Oh, I think this works out bad for NVIDIA.”
Listen at 11:57
U.S. officials may have delayed disclosure to oust Jensen Huang.
“Were they sitting on this waiting for the right time to drop this and get Jensen ousted?”
Listen at 13:01
Jensen Huang’s departure will create a power and cultural vacuum at NVIDIA.
“There is a power and cultural vacuum that's got to come away from this business.”
Listen at 16:08
TSMC may prioritize TPU orders over NVIDIA orders.
“I think there might be some more TPU orders. Like, sorry, you're at the back of the line now.”
Listen at 17:27
The controversy probably will not reduce NVIDIA GPU orders soon.
“is this going to reduce GPU orders in the short term? I don't think so.”
Listen at 17:51
NVIDIA’s valuation implies $400 billion of earnings next year.
“that implies $400 billion in earnings next year.”
Listen at 18:25
AMD is likely the primary beneficiary of NVIDIA’s troubles.
“AMD, I think, number one.”
Listen at 19:00
Airbnb’s booking volume is growing above 20%.
“Airbnb is growing at above 20%.”
Listen at 25:14
Airbnb’s stock is not currently cheap.
“I kind of look at it as the stock is not too cheap right now.”
Listen at 27:03
Booking is probably a better investment opportunity than Airbnb.
“I kind of think booking is probably a much better opportunity from here.”
Listen at 27:29
Airbnb’s competitive moat continues to expand.
“the moat continues to expand”
Listen at 28:48
Airbnb’s improved AI search could strengthen differentiation.
“if they get it right, it could really help with the differentiation.”
Listen at 30:31
Booking is preferable because its valuation gap versus Airbnb is excessive.
“probably booking. Booking is... God, that valuation gap is just getting... It's too extreme.”
Listen at 30:53
MercadoLibre trades at its lowest historical gross-profit multiple, 4.7.
“at their lowest multiple of gross profit in history, 4.7.”
Listen at 36:37
MercadoLibre could trade near ten times earnings if margins recover.
“if you think the margin can return, you're probably getting close to 10 times earnings for a business growing pretty quickly.”
Listen at 36:57
MercadoLibre offers unusually asymmetric upside.
“Mercado Libre I think is maybe one of the most asymmetric opportunities where the upside is massive, honestly.”
Listen at 37:52
MercadoLibre should retain durable growth prospects over the next decade.
“they should have durable growth prospects similar to how like, okay, like 10 years from now”
Listen at 39:57
Evan Spiegel has presided over severely underperforming Snap.
“despite outrageous underperformance”
Listen at 41:32
Snap’s shares outstanding have grown 4% annually since 2017.
“since 2017, the shares outstanding have grown at a 4% annual rate.”
Listen at 45:18
CoreWeave and Nebius face unsustainable cash burn.
“there's just going to be a huge cash burn here, and I don't see what's sustainable.”
Listen at 48:17
CoreWeave and Nebius rely on customer prepayments as a financing strategy.
“their grand plan, is to have customers prepay.”
Listen at 48:28
CoreWeave spends roughly 25% of revenue on quarterly interest expense.
“CoreWeave is spending 25% roughly of its revenue on interest expense every quarter.”
Listen at 49:48
CoreWeave’s financing will slow or become materially more expensive.
“that financing is definitely going to slow down for them or they're going to have to take on horrendous terms”
Listen at 50:19
NVIDIA revenue will probably decline within four to five years.
“there's probably a pretty good chance in four to five years revenue is declining.”
Listen at 54:27
NVIDIA’s current GPU demand reflects a finite infrastructure buildout.
“It is a build out, not a build forever”
Listen at 54:53
Micron may eventually lose money, making a one-times multiple attractive.
“Maybe I'd buy it one times because the earnings are going to be, they're going to lose money. at some point in the near future.”
Listen at 55:28
Ryan would probably avoid NVIDIA and Micron because they are cyclical.
“I don't know this business super well and it's notoriously cyclical. Boom. That's enough for me to probably avoid these at all costs.”
Listen at 57:37
Remitly could generate $500 million in operating earnings within two to three years.
“I can still very easily see a world where they're generating $500 million in operating earnings within two to three years.”
Listen at 59:06
Remitly’s stock is undervalued.
“I think it's too cheap.”
Listen at 59:48
AST SpaceMobile’s rocket project signals severe launch-capacity constraints.
“the bottleneck's so large, SpaceX is keeping all their launches to themselves, that they decided to build their own rocket. I think that is a sign. Things are not going well.”
Listen at 1:00:34
Rocket Lab’s third-quarter revenue rose 73% and backlog reached $3.6 billion.
“Q3 revenue up 73%, some of it inorganic, backlog up to $3.6 billion.”
Listen at 1:01:54
Rocket Lab becomes attractive around $20–$30, but is expensive near $65.
“I think like $20 to $30 I get interested in buying, but right now we're like $65. It's still very, very expensive.”
Listen at 1:02:49
Rocket Lab previously traded at 100 times sales despite low margins.
“At some point, it was at 100 times sales for a low-margin business.”
Listen at 1:03:12
Space businesses are difficult to build during bear markets because they require cheap capital.
“these are businesses that would be hard to create in a bear market. You need capital to do it, and you need cheap capital.”
Listen at 1:04:32
Statements are attributed to the speaker as said on the episode and reflect their view at the time, not PodLume's. They are not advice.
Weekly email
State of the Pod, every Monday
The week's most-discussed names and stories from podcasts, with the moments behind them. Free, five minutes, unsubscribe in one click.