• Consider Broadcom (AVGO) for exposure to custom AI-chip demand; one investor bought at $339.45, while a hypothetical valuation scenario—not a forecast—suggested roughly $600.
• Nebius (NBIS) may be a higher-conviction dip-buy than CoreWeave (CRWV): one investor began a long-term position at $182, but wait for evidence of stable financing and AI demand rather than assuming the selloff has ended.
• TSMC (TSM) offers exposure to a leading chip manufacturer with strong earnings; one investor bought at $420.31, though semiconductor volatility remains a risk.
• Avoid treating unconfirmed AI-infrastructure reports or short-term rebounds as signals; CoreWeave’s debt and profitability concerns remain unresolved.

CoreWeave (CRWV)

• CoreWeave was a focus of the selloff and rebound. Its shares fell more than 22% during the week at one point, then rose about 5% on the day discussed, closing near $69.
• The transcript links investor concerns to rising yields on CoreWeave’s debt and questions about whether its business can generate enough profit to service that debt. The speakers said they did not know who was selling the bonds or whether the selling reflected company-specific concerns.
• Nvidia CEO Jensen Huang said CoreWeave’s strength was improving, its cost of capital was decreasing, and Nvidia did not need to backstop it. Those comments helped ease fears, but did not resolve questions about the debt.
• The guest described CoreWeave as the weakest-positioned of the “neocloud” businesses and said he would not buy it, citing concerns about its business model, margins, and debt. The host, by contrast, made a short-term trade in the stock and said he did not hold the position.

Takeaways

• Treat the rebound as uncertain: the speakers disagreed on whether it marked a bottom or a temporary relief rally.
• For a long-term thesis, monitor debt costs, profitability, and the company’s ability to meet its commitments; the discussion identified these as key unresolved issues.

Nebius (NBIS)

• The host said Nebius had fallen from around $250 the prior week to near $160 intraday, then recovered to around $188.
• The host bought $1,000 at $182 in a retirement account, viewing the decline as an opportunity to begin adding to a long-term position.
• The guest said Nebius appeared to be in a better position than CoreWeave, though he cautioned that neocloud debt and the broader AI spending cycle remained important risks.
• Nebius was also mentioned among the companies held by a fund that was rumored to be under stress. The transcript said the rumor was unconfirmed and that sources disputed it.

Takeaways

• The host’s purchase reflects a willingness to buy after a steep drop, not confirmation that the stock had bottomed.
• Watch financing conditions and AI customer demand, and distinguish verified company news from unconfirmed market rumors.

Broadcom (AVGO)

• The host bought $1,000 at $339.45 in a retirement account.
• The guest was more positive on Broadcom than Nvidia for potential relative performance over the following year. He argued that demand for custom AI chips, or ASICs, could give Broadcom room to gain share.
• The guest cited a projected 2028 EPS of $30 and illustrated that applying a 20× multiple would imply a share price near $600. This was a hypothetical valuation example, not a stated company target.
• The discussion also noted that OpenAI and Anthropic are pursuing custom chips, partly to reduce dependence on expensive Nvidia GPUs.

Takeaways

• The potential opportunity described is exposure to custom AI-chip demand, but the thesis depends on customers continuing to build and adopt those chips.
• Treat the guest’s valuation example as a scenario, not a forecast; the discussion did not establish that Broadcom’s projected earnings or multiple are assured.

Nvidia (NVDA)

• Nvidia shares recovered from an intraday low near $210 to close around $217, but did not finish green.
• Jensen Huang’s remarks that CoreWeave was improving and did not need an Nvidia backstop helped calm concerns about the neocloud sector.
• The guest said Nvidia could continue to perform well but argued that it has more market share to lose than Broadcom. He also raised the risk that large customers may develop their own chips to reduce their dependence on Nvidia and its pricing.
• The host pushed back on the idea that Nvidia would lose meaningful share, noting that custom chips may not suit every workload.

Takeaways

• The discussion presents Nvidia as a leading AI-chip business with continued demand, but facing possible competitive pressure from custom silicon and customer efforts to improve economics.
• Track both overall AI spending and the portion captured by Nvidia; the guest argued that Nvidia’s growth may not keep pace with total AI spending even if its revenue continues to rise.

Micron Technology (MU)

• Micron recovered from an intraday low near $858 and finished around $927.
• The host cited an unconfirmed report that Google was discussing memory supply agreements through 2035. The same discussion said HP and Dell were extending supply agreements into 2031, and that shifting memory capacity toward data centers could reduce PC and smartphone supply by 20%–30% next year.
• The guest bought options on Micron after seeing an intraday reversal and the memory-contract report. He argued that longer-term agreements could extend the memory cycle and make Micron’s future cash flows more durable.
• The speakers cautioned that the thesis depends on hyperscalers’ future spending and on memory demand and pricing.

Takeaways

• The potential bullish case is that sustained data-center demand supports memory sales for longer than a typical cycle.
• The Google agreement was discussed as a report rather than a confirmed company announcement; watch for confirmation and for hyperscaler commentary on spending.

Super Micro Computer (SMCI)

• The host bought $1,000 at $34.27, describing the shares as oversold and saying he believed the stock deserved to be higher. He did not give detailed supporting analysis in the discussion.
• SMCI was part of the broader group of AI and semiconductor stocks that fell sharply and then partially recovered.

Takeaways

• The host’s purchase was a dip-buying decision, not evidence that the decline was over.
• The transcript offered limited company-specific analysis, so the broader AI and semiconductor risks discussed should not be mistaken for a detailed SMCI investment case.

Taiwan Semiconductor Manufacturing Company (TSMC) (TSM)

• The host bought $1,500 at $420.31, citing TSMC’s strong earnings.
• The shares were down about 2% on the volatile day discussed, despite the earnings report.
• The guest praised TSMC’s approach to maintaining long-term customer relationships rather than maximizing prices, using it as a contrast to his concerns about Nvidia’s pricing.

Takeaways

• The discussion was broadly positive on TSMC’s earnings and strategic position, but the stock did not escape the market-wide volatility.
• Consider the company’s earnings alongside the broader semiconductor cycle and customer demand.

Advanced Micro Devices (AMD)

• AMD recovered during the session, moving from around $530 to $565, but finished lower on the day.
• The guest said AMD could benefit from growing demand for CPUs as AI agents require more computing resources, but acknowledged that he had not studied AMD as closely as Nvidia or Broadcom.
• He also characterized AMD’s valuation as high, citing a 150× P/E in the discussion.

Takeaways

• The potential opportunity discussed is CPU demand associated with AI agents, but the guest’s view was less developed than his analysis of Broadcom and Nvidia.
• The transcript did not provide a price target or a specific recommendation for AMD.

Semiconductor Exposure: VanEck Semiconductor ETF (SMH) and AI-Related Hardware

• SMH fell as much as about 4% intraday, then recovered to a smaller loss by the close.
• The host bought $1,000 of SMH? The transcript lists a purchase of SMCI and does not list an SMH purchase; SMH was discussed as a market indicator.
• The speakers described the semiconductor selloff as potentially connected to questions about AI spending, financing costs, and whether AI revenues justify the scale of planned capital expenditure.
• They also warned that a sharp decline in major AI and semiconductor stocks could weigh on index funds and ETFs because of the sector’s large market weight.

Takeaways

• SMH provides broad semiconductor exposure, but the discussion emphasized that sector-wide declines can be sharp and may reflect changing expectations for AI spending.
• Avoid treating a single-day recovery as evidence that financing or demand concerns have been resolved.

SoFi Technologies (SOFI)

• The host bought $1,000 at $14.78 in a retirement account, describing SoFi as a long-term holding with sustainable earnings potential.
• The stock closed above $15, up about 6% on the day.
• The transcript did not provide detailed company-specific financial analysis.

Takeaways

• The host’s view was long-term and based on expected earnings power, but the discussion supplied little evidence to assess that thesis independently.
• The short-term rebound does not by itself establish a change in SoFi’s fundamentals.

Tesla (TSLA)

• The host bought $1,000 at $361.74, saying he liked Tesla as a long-term business despite concerns about its valuation.
• The shares closed near $362. The discussion also focused on a planned Roadster event, with the speakers expressing interest in whether the demonstration would show something distinctive.
• Brad Gerstner of Altimeter said his firm had moved part of its portfolio into Meta and Tesla; the transcript said Altimeter had sold most of its Tesla position earlier in the year and had initiated a new position.
• The host said he believed Tesla could be a long-term winner but did not like its valuation.

Takeaways

• The discussion highlighted a tension between confidence in Tesla’s long-term prospects and concern about the price investors are paying.
• Treat the Roadster event as a potential news catalyst, not evidence of future sales or earnings.

Rocket Lab (RKLB)

• The host bought $1,000 at $53.44, describing Rocket Lab as a long-term holding with sustainable earnings potential.
• The stock closed near $55.25, still down about 6% on the day.
• The transcript did not provide detailed company-specific financial analysis.

Takeaways

• The host’s purchase was a long-term dip-buying decision, but the discussion did not establish a specific valuation case or price target.
• Consider company fundamentals separately from the broader market rebound.

Meta Platforms (META) and Apple (AAPL)

• Meta shares rose after news that Meta and Apple were partnering. The discussion said Meta’s MuseSpark open-weight model would run on Mac Studios.
• The speakers speculated that the companies might collaborate further, potentially involving Meta’s models and Apple’s products, but presented this as a possibility rather than a confirmed arrangement.
• Brad Gerstner said Altimeter had moved 10% of its book into Meta and Tesla and had held Meta for a long time.
• Apple ended the day lower, despite the partnership news.

Takeaways

• The partnership could create opportunities for Meta’s AI models and Apple’s hardware, but the discussion did not confirm that Meta would power Apple’s assistant or become a default model.
• Follow up on any formal product or commercial announcements before drawing conclusions about revenue impact.

Anthropic and OpenAI (Private AI Companies)

• The discussion focused on whether AI companies’ revenues can justify the scale of planned infrastructure spending.
• Brad Gerstner said reported expectations for Anthropic had risen substantially, while also describing concerns about slowing sequential growth and whether open-source models could commoditize frontier AI services.
• A later report discussed in the transcript said some Anthropic investors were weighing an IPO valuation of $1.5 trillion, compared with earlier talk of $2 trillion. The speakers viewed a lower valuation as potentially more reasonable, while noting that it could affect how much capital the company raises.
• The speakers said future revenue growth, customer spending, and the ability to make AI infrastructure profitable remain uncertain.

Takeaways

• If considering any future IPO, assess the valuation against demonstrated revenue, growth, and the company’s capital needs rather than relying on projections alone.
• The transcript did not provide a confirmed IPO valuation or specific recommendation to invest.

Other AI Infrastructure Companies: IREN, Applied Digital (APLD), Bloom Energy (BE), Hut 8 (HUT), and Nscale

• IREN and Applied Digital were mentioned among companies affected by the broader AI-infrastructure selloff. Applied Digital was roughly flat on the day after earlier losses.
• Bloom Energy recovered during the session. The speakers noted that Bloom was expected to enter the S&P 500, which could attract index-related flows.
• Hut 8 was named among the holdings of a fund involved in unconfirmed market rumors.
• Nscale, a private data-center company, was reported to have encountered construction delays, cooling-system leaks, regulatory hurdles, and a fire. The transcript said only 6.4 MW of a planned 30 MW facility was operational, raising questions about its ability to execute its expansion plans and proceed toward an IPO.
• The speakers also discussed local opposition to data centers as a potential constraint on construction, even where companies have approvals.

Takeaways

• The AI-infrastructure opportunity depends not only on demand, but also on financing, construction execution, and the ability to secure approvals and community support.
• Distinguish listed companies from private businesses such as Nscale, and treat the fund-related rumor as unverified.

Real Estate and Housing: Zillow (Z), Opendoor (OPEN), and Home Depot (HD)

• Zillow and Opendoor rose as Treasury yields fell during the session; Home Depot also gained.
• The host said Zillow’s revenue was growing 18% year over year despite a weak housing market. He identified stock-based compensation and declining year-over-year traffic as concerns, while attributing the traffic weakness partly to the housing market.
• The guest said lower interest rates could help housing-related companies by allowing more people to buy homes, and connected the sector’s rebound to the decline in yields.

Takeaways

• The discussion’s bullish case for housing-related stocks depends in part on lower rates and a recovery in housing activity.
• For Zillow, the host specifically flagged stock-based compensation and declining traffic as issues to monitor.

Interactive Brokers (IBKR)

• Interactive Brokers reported quarterly EPS of $0.67, one cent below the cited analyst expectation, while revenue of $1.9 billion beat the cited estimate of $1.85 billion.
• The transcript reported revenue growth of 15% year over year, net interest income growth of 19%, commissions growth of 12%, and daily average trades growth of 18%.
• Margin loans rose 36% to $105 billion, accounts increased 35% to 5.58 million, and customer equity rose 27% to $964 billion.
• The stock was roughly flat after hours. The host described the results as unremarkable overall, while noting the growth in trading activity and margin loans.

Takeaways

• The reported figures showed strong growth in accounts, customer equity, trading, and margin loans, alongside a slight EPS miss.
• The increase in margin loans is an important indicator of investor activity, but the transcript did not discuss the associated credit risks in detail.

Market Indexes and Bonds: S&P 500 (SPY), Equal-Weight S&P 500 ETF (RSP), and 20+ Year Treasury Bond ETF (TLT)

• The S&P 500 ended slightly higher, near 772, and remained close to its all-time highs even as many semiconductor and AI-related stocks had fallen sharply.
• The guest suggested a more balanced approach: exposure to the equal-weight S&P 500, bonds, and a partial hedge against the S&P 500. He gave a hypothetical allocation of 60% RSP, 40% TLT, and a short position equal to about 33% of the S&P 500.
• He also said a gradual slowdown in AI capital spending would be less damaging than a sudden collapse, because investment could rotate into other sectors and bonds.
• The speakers linked bond yields to AI infrastructure financing and discussed inflation and the Iran war as additional factors affecting rates.

Takeaways

• The guest’s portfolio outline was a defensive example, not a universal recommendation; it involved hedging and bond exposure to manage AI-related volatility.
• The transcript stressed uncertainty around both AI spending and interest rates, so the proposed mix should be understood in that context.

Other Stocks Mentioned

• Robinhood (HOOD): Rose about 3% on the day. The host noted that Interactive Brokers’ results could serve as a comparison for the retail-investing business.
• Palantir (PLTR): Closed above $200; no company-specific thesis was discussed.
• ServiceNow (NOW), Salesforce (CRM), and Akamai (AKAM): Recovered during the session; no detailed investment thesis was provided.
• Dell (DELL), Intel (INTC), ASML (ASML), and Reddit (RDDT): Had notable price moves during the day, but the discussion did not establish specific recommendations.
• Visa (V) and Mastercard (MA): Were described as near all-time highs. The speakers grouped them with established businesses generating strong free cash flow.
• McDonald’s (MCD), Coca-Cola (KO), and Nike (NKE): Were cited as examples of established or “boring” companies that performed well during the session’s rotation away from AI-related stocks.
• Uber (UBER), Grab (GRAB), Zeta (ZETA), Netflix (NFLX), ELF Beauty (ELF), and Wingstop (WING): Were mentioned in connection with their daily price moves; no detailed investment case was provided.
• D-Wave Quantum (QBTS) and IonQ (IONQ): Were mentioned as AI-related companies that benefited from the broader news flow, without further analysis.
• Seagate (STX), Western Digital (WDC), Celestica (CLS), Coherent (COHR), Comfort Systems (FIX), Lumentum (LITE), and Verdant: Were named among the holdings of a fund that was rumored to be experiencing problems. The transcript said the rumor was unconfirmed, so it should not be treated as verified company news.

Takeaways

• Most of these stocks were cited only for price action or as examples; the transcript did not provide enough company-specific analysis to support an investment conclusion.
• The reported fund rumor was explicitly uncertain, and the speakers said sources disputed claims that the fund was in trouble.

Bitcoin (BTC)

• Bitcoin was mentioned only briefly: the speakers said its rebound helped support Robinhood’s shares.
• No Bitcoin price, specific outlook, or direct recommendation was provided.

Takeaways

• The transcript offered no standalone investment thesis for Bitcoin; its only connection was as a possible influence on Robinhood’s trading activity.