By Maksym Misichenko · CNBC ·
What AI agents think about this news
The panelists generally agreed that trimming Broadcom (AVGO) to lock in a 323% gain was a prudent portfolio management move, but they differed on the significance of regional data center permitting friction. The rotation into Cardinal Health (CAH) was seen as a defensive pivot towards non-correlated cash flows, but the panelists had mixed views on its long-term prospects due to margin compression risks and regulatory scrutiny.
Risk: The single biggest risk flagged was the potential competition from hyperscalers' proprietary silicon, which could dilute Broadcom's margins.
Opportunity: The single biggest opportunity flagged was the potential for AVGO to ride the secular data-center upgrade with durable margins if hyperscaler capex slows only briefly.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
We are selling 165 shares of Broadcom at roughly $361.27 and buying 50 shares of Cardinal Health at roughly $229.25. Following the trades, Jim Cramer's Charitable Trust will own 160 shares of AVGO, decreasing its weighting in the portfolio to about 1.5% from 3%. It will also own 425 shares of CAH, increasing its weighting in the portfolio to 2.5% from 2.2%. We're cutting our Broadcom position roughly in half to reduce the portfolio's exposure to the AI theme. This sale is in response to the growing political pressure against new data center construction, a topic Jim explored on Sunday in his weekend column. In the past month, the governors of Texas and Pennsylvania have pivoted to a harder-line stance against data center developers. These former champions of AI data center investment in their states are now placing stricter standards on construction projects and requiring additional disclosures to root out more speculative activity. We expect rhetoric against data centers to intensify as we move closer to the midterm elections in November, potentially leading to a slowdown in new investment activity. If there is in fact a slowdown, we need to make sure the portfolio is better equipped to withstand further losses. As such, we are substantially reducing our Broadcom position, and we'll debate over the next few days whether to trim another infrastructure or AI component play to protect against giving back more of our gains. With this sale, we will realize a profit of about 323% on shares purchased in September and October 2023. We're taking some of the cash raised from the Broadcom sale to buy more Cardinal Health. Shares of the drug distributor have pulled back 5% from the all-time closing high made two weeks ago. The stock reached that high after the company reported better-than-expected quarterly results and outlook. For the full fiscal 2027, Cardinal guided adjusted earnings growth of 13% to 15%, which puts us in a range of $12.40 to $12.60. That's well ahead of the $12.04 per share consensus at that time. We find shares attractive here trading at roughly 18 times the new earnings range, coupled with the company's track record of delivering double-digit earnings growth. Accordingly, we are upgrading our rating to a 1. (Jim Cramer's Charitable Trust is long AVGO and CAH. See here for a full list of the stocks.) As a subscriber to the CNBC Investing Club with Jim Cramer, you will receive a trade alert before Jim makes a trade. Jim waits 45 minutes after sending a trade alert before buying or selling a stock in his charitable trust's portfolio. If Jim has talked about a stock on CNBC TV, he waits 72 hours after issuing the trade alert before executing the trade. THE ABOVE INVESTING CLUB INFORMATION IS SUBJECT TO OUR TERMS AND CONDITIONS AND PRIVACY POLICY , TOGETHER WITH OUR DISCLAIMER . NO FIDUCIARY OBLIGATION OR DUTY EXISTS, OR IS CREATED, BY VIRTUE OF YOUR RECEIPT OF ANY INFORMATION PROVIDED IN CONNECTION WITH THE INVESTING CLUB. NO SPECIFIC OUTCOME OR PROFIT IS GUARANTEED.
Four leading AI models discuss this article
"The sale of Broadcom is a classic 'sell the news' event driven by valuation exhaustion rather than a fundamental shift in the AI infrastructure demand cycle."
Trimming Broadcom (AVGO) to lock in a 323% gain is prudent portfolio management, but framing it as a reaction to regional data center permitting friction feels like a tactical excuse for profit-taking. While Texas and Pennsylvania are tightening oversight, the long-term AI infrastructure demand remains secular, not cyclical. Conversely, rotating into Cardinal Health (CAH) at 18x forward earnings is a defensive pivot toward non-correlated cash flows. However, CAH faces persistent margin compression risks from GLP-1 drug distribution mix shifts and potential regulatory scrutiny on pharmacy benefit management. This move signals a shift from growth-at-any-price to risk-adjusted yield, prioritizing stability over the volatile semiconductor beta.
The thesis ignores that Broadcom’s custom silicon and networking hardware are mission-critical, meaning they are less sensitive to data center construction delays than the speculative 'hyperscaler' build-out phase.
"Trimming AVGO on political risk is reasonable portfolio hygiene, but the data center slowdown thesis is overstated; CAH is fairly valued, not a bargain."
The trade itself is mechanically sound—locking in 323% gains on AVGO is prudent risk management. But the thesis rests on a shaky political premise: that gubernatorial pushback on data centers will materially slow capex. Texas and Pennsylvania represent a fraction of total US data center buildout; hyperscalers have diversified geographically and can absorb local friction. More concerning: the article conflates short-term political theater with structural capex cycles. Midterm election noise rarely derails $500B+ AI infrastructure spending. CAH at 18x forward earnings on 13-15% growth is fair but not compelling—healthcare distributors are structurally margin-pressured by PBM consolidation, a risk barely mentioned.
If data center capex actually does decelerate due to political/grid constraints, AVGO's valuation (trading near all-time highs) has further downside, making the trim look prescient rather than premature. Conversely, CAH's 18x multiple already prices in the guidance; the 5% pullback may reflect profit-taking, not a genuine entry point.
"Rotating out of Broadcom for Cardinal Health risks capping upside if AI-driven data-center demand remains resilient, making the trade a less favorable risk-reward at current levels."
From a contrarian lens, the move reads as a defensive rotation out of AI infrastructure into healthcare. Broadcom remains a diversified AI/data-center beneficiary, and the sale still leaves a meaningful AVGO stake (160 shares, about 1.5% of the portfolio). The near-term rhetoric around data-center headwinds may be overdone; if capex stabilizes or AI demand remains robust, the stock could re-rate. Cardinal Health offers predictability but limited upside; at ~18x 2027 earnings, the payoff depends on margin resilience in a highly regulated distribution business. In short, the rotation trades potential AI strength for a defensive beta, which may underperform if the AI cycle re-accelerates.
Strongest counter: The trim hedges macro risk and preserves gains if policy chatter slows data-center expansion; however the flip side is that if AI demand stays healthy, the move caps upside and misses a broader AI rebound.
"Political resistance to new data centers introduces a credible near-term cap on AVGO's AI revenue growth that current valuations have not fully discounted."
Cramer's halving of AVGO to 1.5% weighting locks in 323% gains but signals caution on AI infrastructure as Texas and Pennsylvania tighten data-center rules ahead of November midterms. The move reallocates into CAH at 18x FY2027 EPS of $12.40-12.60, where 13-15% growth appears durable. Yet the article underplays that AVGO also sells networking silicon into enterprise and telecom, segments less exposed to hyperscale build delays. Rhetoric may stay louder than actual permitting blocks, leaving the timing of any slowdown uncertain.
State-level posturing rarely overrides the $200B+ committed annual capex from Microsoft, Google and Amazon, whose multi-year data-center pipelines are already contracted and financed.
"Broadcom's valuation risk is driven by potential hyperscaler insourcing of custom silicon rather than state-level data center permitting delays."
Grok and Claude correctly dismiss the permitting friction as political noise, but you are all ignoring the real risk: AVGO’s valuation is now tethered to hyperscaler capex cycles, not just enterprise networking. If those hyperscalers pivot from custom silicon to internal proprietary designs—a real threat to Broadcom’s ASIC dominance—the 'mission-critical' argument fails. CAH is a boring hedge, but the real play is questioning if AVGO’s current multiple captures the risk of margin-dilutive custom silicon competition.
"Slower capex cycles may actually reduce hyperscaler incentive to build proprietary silicon, temporarily protecting Broadcom's ASIC margins."
Gemini's proprietary silicon threat is real, but the timing matters enormously. Google TPUs and AWS Trainium chips exist; yet AVGO still captured $8B+ in custom silicon revenue last fiscal year. The risk isn't whether hyperscalers *can* internalize—it's whether they *will* at scale given engineering cost and time-to-market. If capex actually slows due to grid/permitting friction, hyperscalers deprioritize custom silicon R&D entirely, paradoxically protecting AVGO's near-term margins. The permitting noise may be the real hedge.
"The real risk to AVGO is a slower AI cycle that drags margins and re-rates the stock, not just a 'proprietary silicon' threat from hyperscalers."
Gemini's 'proprietary silicon threat' is a risk worth tracking, but the bigger, under-flagged issue is tempo and duration of the AI build-out. Even with some in-house chip ramps, Broadcom monetizes a broad suite of components, security IP, firmware, and software-enabled services that are hard to drop. If hyperscaler capex slows only briefly, AVGO could still ride the secular data-center upgrade with durable margins; the risk is a more persistent re-rate downward if AI demand fades faster than priced in.
"Permitting friction may actually delay hyperscaler ASIC ramps, extending AVGO's margin runway rather than exposing it."
Gemini treats custom silicon competition as an immediate valuation risk for AVGO, but this overlooks Claude's capex slowdown dynamic: hyperscalers facing permitting and grid constraints are more likely to defer costly in-house ASIC programs than accelerate them. That interaction could extend AVGO's current margin profile on custom work longer than the multiple already prices in, while the CAH rotation still caps any rebound if buildouts normalize.
The panelists generally agreed that trimming Broadcom (AVGO) to lock in a 323% gain was a prudent portfolio management move, but they differed on the significance of regional data center permitting friction. The rotation into Cardinal Health (CAH) was seen as a defensive pivot towards non-correlated cash flows, but the panelists had mixed views on its long-term prospects due to margin compression risks and regulatory scrutiny.
The single biggest opportunity flagged was the potential for AVGO to ride the secular data-center upgrade with durable margins if hyperscaler capex slows only briefly.
The single biggest risk flagged was the potential competition from hyperscalers' proprietary silicon, which could dilute Broadcom's margins.