Oct 21, 2026 · 44 min
Big Tech earnings expose AI’s cost and valuation test
Big Tech Earnings Filter In… And Coreweave’s Tangled Web 10/21/26
The episode connects corporate AI ambitions with rising compute costs, heavy financing needs, concentrated customers, and weakening consumer and housing signals.
- 1Meta’s AI monetization story is advancing, but Muse adoption and rising compute costs leave its capital needs unresolved.
- 2Tesla’s earnings beat meets ambitious autonomy and chip targets, while SpaceX speculation complicates how investors value the company.
- 3CoreWeave’s debt and customer concentration show how infrastructure risks can spread through the broader AI trade.
Don't miss
The panel connects CoreWeave’s debt and customer concentration to broader risks in the AI infrastructure trade.
The brief
Meta’s earnings put AI monetization and advertising gains alongside a harder question: whether rising compute and capital costs will outrun the business case.
Tesla delivered an earnings beat, but the market focused on ambitious autonomy and chip targets—and on how a possible SpaceX transaction could reshape valuation.
Gene Munster weighs whether Meta has explained Muse’s path to monetization, as retailer adoption, user engagement, competing AI agents, and funding needs remain unsettled.
IBM becomes the enterprise-AI counterpoint, while Chipotle’s weakening sales and price increases test whether consumer pressure can overwhelm a still-demanding valuation.
CoreWeave’s debt burden and customer concentration sharpen the infrastructure risk, before housing data adds a separate warning about rates, demand, and margins.
The standout tension is an AI market promising rapid expansion while its companies face escalating compute bills, concentrated customers, and uncertain capital requirements.
What was said on this episode
38 statements · 16 positive · 19 negative · 3 neutral
Transaction fees would give Meta a new revenue direction beyond advertising
“taking a cut of transactions is really a different direction for Meta and for Mark Zuckerberg”
Listen at 3:18
Buying Muse at current Meta levels may be premature
“I think it might be a little bit premature, especially at these levels”
Listen at 5:16
Meta is signaling that it will need to raise substantially more cash
“they're signaling that they're going to have to raise a lot more cash”
Listen at 5:36
Meta may be free-cash-flow negative for the first time in roughly ten years
“For the first time now, I think in 10 years, free cash flow negative for Facebook”
Listen at 6:32
Meta’s valuation is reasonable
“Valuation is fine”
Listen at 6:44
Investors should maintain limited exposure to Meta at current levels
“I'd travel lightly here in the name right at these levels”
Listen at 6:51
Meta’s latest results represent largely more of the same
“I think it's a lot more of the same”
Listen at 7:01
Meta’s underlying advertising business had a very strong quarter
“for the underlying part of the business, which is funding this, I thought this was a really good quarter”
Listen at 8:00
Meta gained approximately $500 billion in market capitalization over three months
“it's gained $500 billion in market cap in the last 3 months”
Listen at 9:56
A Tesla valuation around $420 per share is plausible given Musk’s history
“I don't think $420 is ridiculous given his history”
Listen at 13:16
Muse will be central to Meta’s future
“Muse is the future of the company”
Listen at 14:17
Muse likely has approximately ten million daily users
“we're probably about 10 million daily users of Muse”
Listen at 15:08
Personalized bots will have an impact comparable to generative AI chatbots
“the personalized bot is going to have that impact”
Listen at 16:13
Meta will likely raise its next-year CapEx guidance above $225 billion
“I bet when they report December, they probably take it to above $225 billion”
Listen at 16:32
Marketing will increasingly occur between bots rather than directly to humans
“The future of marketing is not to humans, it's bot to bot”
Listen at 18:15
Multiple billions of people will use personal bots within three years
“We're going to have multiple billions of people using these personal bots in the next 3 years”
Listen at 18:24
Meta will need to raise roughly $25–50 billion in capital
“they're going to have to raise $25, $50 billion, something like that”
Listen at 19:00
AI investments should not be expected to pay off within one year
“I don't think you can expect to be paid out in the next quarter or even the next year”
Listen at 20:02
AI investment opportunities require a very long-term investment horizon
“I think this is a very long-term play”
Listen at 20:06
IBM shares can continue rallying from current levels
“I still think this stock sort of can rally from here for sure”
Listen at 23:13
IBM is among the least interesting ways to express enterprise AI adoption
“this is probably one of the most least interesting ones”
Listen at 23:50
Chipotle is relatively expensive among fast-food restaurants and is underperforming
“I would consider Chipotle a bit more expensive on the, on the fast food front and it's not doing as well”
Listen at 26:19
Chipotle’s upcoming quarter appears likely to disappoint
“it feels as though this quarter is setting up for another disappointment”
Listen at 27:06
Investors should wait for Chipotle’s upswing rather than attempt to buy its bottom
“I'd rather buy it higher, even meaningfully higher on an upswing rather than try to pick the bottom”
Listen at 28:20
SpaceX is unlikely to reach ten gigawatts of compute by end-2027
“the likelihood of that is not particularly great”
Listen at 30:39
SpaceX reaching three or four gigawatts could hurt concentrated NeoCloud companies
“that might be really negative for these companies that, again, have a lot of customer concentration”
Listen at 30:47
CoreWeave distress could create ripple effects across the AI infrastructure sector
“there would be ripple effects across all the things that you think are very safe in the space”
Listen at 31:37
CoreWeave shares are likely to experience a tradable rebound
“I still think there's a tradable bounce coming”
Listen at 32:03
CoreWeave’s 9⅝% debt yield is approaching 15%
“the yield is now approaching 15%”
Listen at 32:49
Technology changes may undermine the long-term value of current AI infrastructure spending
“the technology could change and whether the spending today was actually money good, you know, a few years down the line”
Listen at 33:32
FSD will become more important in vehicle purchasing decisions
“FSD is going to become a bigger part of the conversation”
Listen at 35:37
A Tesla–SpaceX transaction has a 95% chance within two years
“I'd put it at 95% chance something happens in the next 2 years”
Listen at 36:24
Tesla’s fundamentals, FSD, vehicle outlook, and Optimus prospects are strong
“if you just look at the fundamentals, the margins, what's going on with FSD, some of the outlook on the vehicles, Optimus comments, things are going very well at Tesla”
Listen at 36:33
Combining SpaceX with Tesla could create a broader, potentially positive investment
“the SpaceX thing maybe makes it a broader play. So maybe that's, maybe that's a positive thing”
Listen at 38:21
Rents are rising again as multifamily supply is absorbed
“rents are coming up again”
Listen at 40:19
Builder concessions and price cuts will pressure homebuilder margins
“How is this going to hit the margins, which are not good?”
Listen at 40:45
Tesla’s CapEx reduction raises questions about its growth outlook
“Why is this such a great growth company? Why are they lowering CapEx?”
Listen at 41:18
TLT has further upside after its recent rebound
“I think there's more room to go to the upside for TLT”
Listen at 43:05
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.