• Favor Microsoft (MSFT) and Amazon (AMZN) as conditional longer-term plays: a slower, more sustainable pace of AI data-center construction could ease capital needs and improve cash flow, potentially by late 2027.
• Be cautious with debt-dependent AI data-center builders CoreWeave (CRWV), Nebius (NBIS), and IREN; rising financing costs could constrain expansion.
• Treat AI infrastructure stocks—including NVIDIA (NVDA), AMD (AMD), and Vertiv (VRT)—as high-risk until the sustainability of data-center spending becomes clearer.
• Avoid Fiserv (FI) and FIS, which the host explicitly advised investors to stay away from amid weak growth and competitive pressures.

Broad U.S. Equities (S&P 500, Nasdaq)

• The market fell sharply on Thursday, with the S&P 500 down 4.4% and the Nasdaq down 5.5%, as investors considered whether the midterm results could lead to slower AI data-center construction.
• The host said AI capital spending accounts for at least 50% of GDP growth this year. A slowdown could slow the economy, though he said it would not necessarily cause a recession.
• The initial selloff was described as panic-driven, with some hedge funds potentially forced to liquidate positions.

Takeaways

• The discussion suggests monitoring policy proposals affecting data-center construction, as well as AI companies’ capital spending and financing needs.
• The host sees the possibility of slower AI investment as both a risk to AI-related growth and a potential route to more sustainable spending by hyperscalers.

AI Data-Center Construction and Infrastructure

• The host listed power suppliers, semiconductor makers, data-center equipment providers, industrial companies, and “neocloud” operators as exposed to a slowdown in data-center construction.
• Stocks in these groups fell sharply during Thursday’s selloff, including Quanta Services (PWR), NVIDIA (NVDA), AMD (AMD), Micron (MU), Dell (DELL), Super Micro Computer (SMCI), Arista Networks (ANET), Cisco (CSCO), Eaton (ETN), Vertiv (VRT), nVent (NVT), CoreWeave (CRWV), Nebius (NBIS), and IREN (IREN).
• The host cautioned that strong quarterly results from companies such as AMD and Arista show current AI spending remains strong, but do not resolve questions about the sustainability of that spending.

Takeaways

• Treat AI-infrastructure companies as sensitive to both data-center demand and the availability and cost of financing.
• The transcript does not identify a single preferred stock in this group; it emphasizes that future policy and capital-spending changes could materially affect growth.

CoreWeave (CRWV)

• CoreWeave’s debt yield reportedly rose from 9% to 15%, a level the host called prohibitive for financing data-center construction.
• He described CoreWeave, Nebius, and IREN as “neoclouds” that rely on borrowing to build data centers and rent them out, unlike hyperscalers with broader businesses to support spending.
• CoreWeave shares fell 9% on Thursday.

Takeaways

• The host’s discussion is bearish on the financing model of neoclouds: high borrowing costs could make expansion difficult.
• Investors following the company may want to watch debt-market conditions and cash-flow needs, not just demand for AI computing.

Nebius (NBIS)

• The host grouped Nebius with CoreWeave and IREN as a neocloud that depends on debt and capital markets to fund data-center construction.
• Nebius shares fell 14.9% on Thursday amid concerns about AI spending and financing.

Takeaways

• The discussion highlights financing dependence as a key risk to Nebius’s growth model.
• The host did not offer a specific price target or recommendation for Nebius.

IREN (IREN)

• IREN was also described as a neocloud dependent on raising capital to build AI data centers.
• Its shares fell 17% on Thursday.

Takeaways

• The transcript points to the same risk identified for other neoclouds: expensive or unavailable financing could constrain data-center growth.
• Consider the company’s reliance on capital markets when assessing its exposure to the AI theme.

NVIDIA (NVDA)

• NVIDIA was listed among the semiconductor companies vulnerable to a slowdown in data-center construction.
• Its shares fell 9.3% on Thursday.

Takeaways

• The transcript frames NVIDIA as exposed to the pace of AI infrastructure investment.
• The key issue raised is not current demand alone, but whether data-center spending can continue at its present pace.

AMD (AMD)

• AMD reported strong results: earnings per share were $2.09, up 74% year over year, and revenue was $13.38 billion, up 45%. The company forecast fourth-quarter revenue of approximately $17 billion.
• Despite the strong report, AMD shares fell 17.3% on Thursday as AI-related stocks sold off.
• The host said good results from chip companies reflect current high levels of AI capital spending but do not settle the broader AI investment debate.

Takeaways

• AMD’s results indicate strong current demand, while the transcript emphasizes that future spending levels and policy changes remain important uncertainties.
• Strong earnings alone may not protect AI-linked stocks if the market reassesses the longer-term spending outlook.

Micron (MU)

• Micron was included among semiconductor companies exposed to a slowdown in AI data-center construction.
• Its shares fell 17.5% on Thursday.

Takeaways

• The discussion ties Micron’s investment outlook to continued AI infrastructure spending.
• The transcript provides no company-specific earnings update or recommendation beyond that exposure.

Dell (DELL)

• Dell was listed among data-center equipment providers that could be affected if construction slows.
• Its shares fell 15% on Thursday.

Takeaways

• The host’s framing makes data-center construction activity a key demand variable for Dell’s AI-related business.
• Monitor whether planned infrastructure projects continue at their current pace.

Super Micro Computer (SMCI)

• Supermicro was included among data-center equipment providers vulnerable to slower construction.
• Its shares fell 8% on Thursday.

Takeaways

• The company’s exposure to AI data-center buildouts makes construction plans and spending trends important to track.
• No specific company guidance or recommendation was discussed.

Arista Networks (ANET)

• Arista reported earnings per share of $1.20, up 60% year over year, and revenue of $3.7 billion, up 58%. Its fourth-quarter guidance was above estimates.
• The host said the results show current AI capital spending remains strong, but do not settle concerns about the future of that spending.
• Arista shares fell 10% on Thursday.

Takeaways

• The earnings report was strong, but the discussion cautions against treating current results as proof that the broader AI investment cycle will continue unchanged.
• Future data-center spending and the policy environment remain central considerations.

Cisco Systems (CSCO)

• Cisco was listed as a data-center equipment provider that could be hurt by slower AI construction.
• It was also described as Arista’s chief competitor.
• Cisco shares fell 7% on Thursday.

Takeaways

• The transcript links Cisco’s AI-related exposure to data-center investment levels.
• It provides no specific company outlook or stock recommendation.

Power and Electrical Infrastructure (GE Vernova, Quanta Services, Eaton, Vertiv, nVent)

• The host identified power suppliers and electrification companies as potential beneficiaries of AI data-center construction, but also as vulnerable if construction slows.
• On Thursday, GE Vernova (GEV) fell 7.5%, Quanta Services (PWR) fell 12.3%, Eaton (ETN) fell 7%, Vertiv (VRT) fell 14%, and nVent (NVT) fell 12%.
• The host said Eaton gets about 20% of revenue from AI-related construction. It reported earnings per share of $3.68, up 20%, and sales of $8.95 billion, up 28%; it also raised 2026 guidance. The host disclosed that he has owned Eaton for some time.

Takeaways

• These companies may benefit from ongoing electrification and data-center investment, but the same exposure creates risk if construction plans slow.
• Eaton’s reported results and raised guidance were positive, but the host also emphasized the possibility of a broader slowdown in AI-related capital spending.

Microsoft (MSFT), Amazon (AMZN), and Alphabet (GOOGL)

• The host suggested that slower AI data-center construction could eventually benefit hyperscalers Microsoft, Amazon, and Google by reducing the need to raise large amounts of capital.
• He said a slower, more sustainable construction pace could help them return to positive cash flow, perhaps as early as late 2027, and allow them to fund AI investment from cash flow.
• During Thursday’s selloff, Microsoft fell 3.6%, Amazon fell 2.2%, and Google fell 1.9%—less than many AI infrastructure stocks.
• The host also noted that Google and Amazon had negative free cash flow of $9 billion and $33 billion, respectively, in the quarter discussed, and said hyperscalers have also been raising capital to fund AI spending.

Takeaways

• The host’s view is relatively constructive on hyperscalers if slower construction improves cash flow and reduces reliance on external financing.
• That potential benefit is conditional: the transcript also stresses that AI spending has already made these businesses more capital-intensive and pressured cash generation.

Alphabet / Google Search (GOOGL)

• Google’s search revenue grew 12%, down from 19% growth in the first quarter of 2026; the host described the result as disappointing.
• He said the concern is that traditional Google Search—a capital-light, high-moat business—could gradually be replaced by AI search, which he described as more capital-intensive and having lower barriers to entry.
• He also noted that switching between AI models appears relatively inexpensive.

Takeaways

• The discussion identifies a possible long-term change to Google’s business mix: less reliance on high-margin search and more investment in capital-intensive AI.
• Track search growth and the economics of AI services, rather than viewing AI adoption as an unqualified benefit to Google.

Meta Platforms (META)

• The host said Meta began a process of raising $90 billion in equity in the last week of October to help fund AI capital spending.
• He argued that even hyperscalers’ AI plans have become partly dependent on capital markets.

Takeaways

• The key issue raised is whether Meta’s AI investment can be supported sustainably without excessive reliance on external capital.
• The transcript does not provide a stock-specific recommendation or describe Meta’s share-price performance.

Oracle (ORCL)

• Oracle began raising $10 billion in debt and equity to help fund its AI data-center construction plans.
• The host said the market had become concerned about Oracle’s balance sheet and need for capital; the stock fell 9% on the announcement.

Takeaways

• The discussion is cautious about Oracle’s financing needs and balance-sheet risk.
• Investors following Oracle may want to watch its capital-raising needs and ability to fund AI projects.

Anthropic and AI Model Providers

• Anthropic’s amended S-1 showed quarterly revenue growth slowing to 47% in the third quarter of 2026, compared with 147% in the second quarter. The host said the earlier surge was partly driven by “token maxing,” when companies encouraged heavy AI use without focusing on cost.
• Anthropic’s revenue-to-compute-expense ratio fell to 1.72 times in the third quarter from 2.08 times in the second quarter.
• The host said Anthropic cut token prices by as much as 50% in October to stimulate demand, and described a broader AI model price war involving Anthropic, OpenAI, and Google’s Gemini.
• He also said Anthropic would continue to lose cash and that the tech correction would hurt its valuation for its planned IPO, which he said was expected in a few weeks.
• The transcript’s figures for Anthropic’s total lease and AI commitments are garbled, so no reliable amount can be extracted.

Takeaways

• The discussion is cautious about AI model providers’ growth quality, compute costs, pricing pressure, and cash burn.
• For a potential Anthropic IPO, the host suggests that slower growth, lower pricing, and a weaker technology market could weigh on valuation.
• OpenAI is discussed as a company and model provider, not as a publicly traded stock in this transcript.

Apple (AAPL)

• Apple reported revenue of $114.5 billion, up 12%, and earnings per share of $2.10, up 14%; both beat expectations.
• Services revenue growth slowed to 10%, and Apple guided to services growth below 10% for the next quarter. The host also noted light margin guidance due to memory-chip price inflation.
• He said Apple is not building its own large language models and therefore retains strong free cash flow. He described its potential strategy of using its iPhone platform to have other AI providers “rent space” as potentially low-risk, high-reward.

Takeaways

• The host views Apple’s platform position and free cash flow as a potential advantage in AI, without requiring the company to build its own models.
• Services growth and memory-chip costs were the main near-term concerns discussed.

Tesla (TSLA) and SpaceX

• The host said Democrats might create roadblocks for companies associated with Elon Musk, though he did not specify particular policy actions.
• On Thursday, Tesla fell 10.8% and SpaceX was reported down 9.1%. SpaceX is not presented as a publicly traded stock in the transcript.
• Separately, SpaceX announced a deal with Qualcomm to provide network technology for Starlink’s mobile service. The host said Musk appeared to be challenging the major mobile providers.

Takeaways

• The transcript identifies potential political opposition to Musk-linked projects as a risk, but does not specify its likely scale or outcome.
• SpaceX’s Starlink expansion could affect the competitive landscape for mobile service, though the transcript provides no investment recommendation on the private company.

Qualcomm (QCOM)

• Qualcomm reported earnings per share of $2.17, down 28% year over year, while revenue was down 8%. Its guidance for the December quarter was slightly below expectations.
• The host attributed weakness to memory-chip price increases hurting handset sales. Qualcomm’s handset-chip sales fell 26% in the quarter.
• SpaceX also announced a deal with Qualcomm for network technology for Starlink’s mobile service.

Takeaways

• The discussion highlights exposure to smartphone demand and memory-chip costs as near-term risks.
• The Starlink agreement is a potential new business opportunity mentioned in the transcript, but no financial impact was specified.

Fiserv (FI) and FIS (FIS)

• Fiserv reported earnings per share of $1.75, above estimates, but revenue of $4.85 billion missed expectations. Organic revenue growth was 0%. Its shares rose 12% after the results, which the host characterized as a relief rally rather than evidence of a long-term turnaround.
• FIS reported earnings per share of $1.61, up 6.6%, and revenue growth of 3.7%. It paused buybacks and some dealmaking to reduce leverage, cut 2026 EPS guidance, and announced that its CFO was stepping down. Its shares fell 24%.
• The host said the payment sector is difficult, with competition and limited barriers to entry outside of Visa and Mastercard. He said Fiserv and FIS have been horrible stocks and concluded, “Best to stay away.”

Takeaways

• The host’s explicit view is to stay away from Fiserv and FIS.
• The risks he cited include competitive pressure, weak or uneven growth, FIS’s leverage, reduced buybacks, and the CFO departure.

Visa (V) and Mastercard (MA)

• The host said that outside of Visa and Mastercard, payment companies generally have few barriers to entry, while competitors increasingly enter each other’s markets.

Takeaways

• The transcript presents Visa and Mastercard as exceptions to the broader payment-sector concerns because of their stronger barriers to entry.
• It does not provide a specific recommendation or valuation view for either company.

Apollo Global Management (APO)

• Apollo reported 23% growth in fee-related earnings, $53 billion of inflows, and 18% growth in assets under management.
• Its spread-related earnings grew only 7% year over year, below expectations. The host said this may reflect a conservative approach to new credit investments.
• Apollo was down 5% on Tuesday and 19% for the year at the time discussed. The host also noted that Apollo is a major financier of AI data centers, making it partly exposed to the AI investment cycle.

Takeaways

• The results were described as mixed: strong fee-related figures but weaker spread-related income.
• Apollo’s exposure to AI financing adds another area to monitor alongside its credit and insurance-related businesses.

Booking Holdings (BKNG)

• Booking Holdings reported earnings per share of $46.60, up 17% year over year and above expectations.
• Revenue was $8.89 billion, below the prior-year figure of $9.01 billion and below estimates. The company guided to fourth-quarter revenue growth of only 0% to 2%.
• The host said investors are concerned that agentic AI could weaken travel brands by emphasizing price, but added that it was too early to see a clear impact. He also pointed to a weakening consumer.

Takeaways

• The transcript flags both consumer weakness and potential longer-term pressure from AI-driven price comparison as issues to watch.
• The earnings and revenue figures point to a mixed near-term picture, rather than a clear conclusion about the effect of agentic AI.

FICO (Fair Isaac Corporation)

• FICO reported revenue growth of 27% and earnings per share of $11.59, up 43% year over year.
• The company did not provide its usual next-year guidance. Management cited competitive pressure from VantageScore and uncertainty around mortgage-scoring pricing.
• The host said FICO had lost its “precious monopoly” and argued that it needed new management.

Takeaways

• The host is bearish on FICO’s competitive position, emphasizing the loss of pricing certainty and the company’s unusual decision not to issue guidance.
• The transcript identifies competition from VantageScore and changes in mortgage scoring as key risks.

T-Mobile (TMUS), AT&T (T), and Verizon (VZ)

• The host said SpaceX and Starlink appeared to be moving toward challenging the three major mobile providers.
• He characterized T-Mobile, AT&T, and Verizon as “uninvestable” in the context of that potential competition.

Takeaways

• The host’s stated view is strongly cautious: he considers the three carriers uninvestable as Starlink’s mobile ambitions develop.
• The transcript does not quantify the competitive threat or give a timeline for any impact.

U.S. Housing

• The host previewed an interview with housing researcher Melody Wright, who believes housing prices will decline in most parts of the country.
• The host described the state of U.S. housing as “horrific,” but this was a preview of the upcoming discussion rather than a detailed analysis of housing-related investments.

Takeaways

• The transcript flags a potentially bearish housing outlook, but provides no specific housing stock, fund, price target, or investment recommendation.