Broadcom (AVGO) on Solana

Broadcom is negotiating a fresh debt package potentially reaching $100 billion, structured as a roughly $60–70 billion senior secured tranche and a $30 billion junior tranche channeled through a special purpose vehicle. The financing is designed to fund AI chip arrangements for customers including Anthropic, extending a June partnership with Apollo Global Management and Blackstone that committed $35 billion to expand Anthropic's computing capacity as part of a push to enable more than 20 gigawatts of AI compute by 2028. Broadcom also signed a five-year deal with Alphabet covering future TPU chip generations and networking equipment, reinforcing its role as the preferred ASIC supplier for hyperscalers seeking custom alternatives to general-purpose GPUs.

The scale of the debt raise tracks Broadcom's accelerating AI revenue base: AI semiconductor sales reached $10.8 billion in the second quarter alone, putting the business on roughly a $43 billion annualized run rate. Management has projected more than $100 billion in AI chip revenue by 2027, a figure that would exceed the company's total 2025 revenue of $64 billion and suggests Broadcom is positioning itself not just as a chip supplier but as a capital intermediary for large-scale AI infrastructure deployment.

Stanley Druckenmiller's Duquesne Family Office and Dan Loeb's Third Point fully exited their Broadcom positions in the most recent 13F filings, selling 195,955 and 50,000 shares respectively, while simultaneously adding to Taiwan Semiconductor. Druckenmiller increased his TSM stake to 589,680 shares, and Third Point lifted its holding to 460,000 shares. David Tepper's Appaloosa ran against the grain by opening a new 150,000-share AVGO position, though Appaloosa's much larger TSM bet — 1.65 million shares worth roughly $788 million — signals the same underlying rotation.

The shift comes despite Broadcom reporting strong Q2 FY2026 results: revenue of $22.2 billion (+48% year-over-year) and AI semiconductor revenue of $10.8 billion (+143%), with Q3 guidance calling for $29.4 billion in revenue including $16 billion in AI silicon. Broadcom's stock is down roughly 8% over the past week and up about 14% year-to-date, while Taiwan Semiconductor has gained 41% in 2026 and 79% over the past year. Analysts note that concerns over customer concentration risk — including reports of Alphabet diversifying its custom-chip suppliers — and 230 basis points of margin compression are weighing on sentiment toward AVGO even as its AI revenue trajectory remains among the strongest in semiconductors.

Broadcom has established an off-balance-sheet chip financing vehicle that could accumulate up to $370 billion in senior debt by mid-2029. The structure allows customers to lease Broadcom's custom AI accelerator chips (XPUs) by backstopping their lease obligations, making it easier for hyperscalers and other large buyers to access Broadcom's AI platforms without committing full upfront capital.

The central risk is what happens if customers default. Because Broadcom effectively guarantees those lease obligations, widespread non-payment could push exposure back onto the company's own balance sheet — meaning the off-balance-sheet label may not hold in a downturn in AI spending. Analysts are flagging future quarterly disclosures on guaranteed lease exposure as the critical metric to watch as the vehicle scales toward 2029, with Broadcom's roughly $2 trillion market capitalization already pricing in a sustained AI infrastructure buildout.

Broadcom (AVGO) currently trades at roughly 71 times trailing twelve-month earnings and 27 times next year's analyst consensus estimates, a 44-point spread that can only resolve in two ways: earnings rise fast enough to justify the price, or the price falls to meet earnings. Historical precedent from Broadcom's own record leans toward the former. In fiscal 2024, the trailing multiple reached 137x against a forward multiple near 29x — a gap that closed as net income doubled and market cap grew from $790 billion to $2 trillion. In fiscal 2021–22, a similar (if smaller) divergence resolved through earnings surges while the stock dipped only around 10%. The narrowing GAAP-to-non-GAAP earnings gap, as VMware acquisition amortization works its way through the income statement, provides an additional mechanical tailwind to reported earnings.

The bull case rests on execution: fiscal Q2 2026 delivered 88% year-over-year net income growth to $9.3 billion, 48% revenue growth to $22.2 billion, and 60% free cash flow growth to $10.3 billion, with Q3 guidance pointing to roughly $29.4 billion in revenue — an 84% YoY increase. If that trajectory holds, the earnings denominator rises fast enough to compress the trailing multiple toward the forward multiple while leaving the stock flat or higher. The bear case is narrower but sharper: at 27x forward, the market has already priced in earnings roughly doubling, so the forward multiple offers little cushion if the ramp slips. Unlike prior wide-gap episodes, any meaningful guidance miss would leave the price side — not the earnings side — as the variable that adjusts.

CEO Hock Tan reiterated Broadcom's target of more than $100 billion in AI semiconductor revenue by fiscal 2027 — a figure that would exceed the company's entire current annual revenue of roughly $75 billion. The trajectory looks credible on paper: Q2 fiscal 2026 AI revenue came in at $10.8 billion, up 143% year-over-year, with Q3 guided above $16 billion — a pace that would imply more than 200% annual growth. Four named customers — Alphabet, Meta Platforms, Anthropic, and OpenAI — anchor the demand, with two undisclosed hyperscalers having placed $6 billion in combined purchase orders for shipments beginning late 2026.

The concentration risk embedded in that forecast is significant. With roughly six customers accounting for nearly the entire AI revenue base, analysts note that averaging $100 billion across six buyers implies roughly $16 billion per customer — and there is no meaningful secondary tier to absorb a miss. A pause or reduction in orders from even one or two of those relationships could materially alter the trajectory of the forecast Tan has staked Broadcom's near-term growth story on.

Broadcom's growth story is becoming harder to reduce to a single headline. While its custom AI chip business posted $10.8 billion in Q2 2026 revenue—up 143% year-over-year—the company simultaneously pushed out a meaningful software security update that signals how it intends to compound that momentum.

On August 6, Broadcom announced upgraded VMware capabilities targeting AI-related cyber threats. The vDefend platform now delivers Distributed Firewall throughput of up to 75Gbps per 100G NIC server, a 241% increase, while the Avi Load Balancer scales to 12.25Tbps per controller. A new two-node deployment model cuts hardware requirements by 33%. The release bundle—vDefend SSP 5.2, vDefend 9.1.1, Avi Load Balancer 32.1.4, and vACT 3.0—adds native API protection and air-gapped support, addressing enterprise security needs that grow in lockstep with AI infrastructure deployments.

The software push matters strategically because it converts the AI chip wave into recurring enterprise infrastructure spend. Hyperscalers building out AI compute also require hardened network security at scale, and Broadcom's VMware stack is positioned to capture that adjacent demand. Full-company Q2 revenue reached $22.2 billion (+48% YoY), adjusted EPS hit $2.44 (+54%), and free cash flow came in at $10.3 billion (+60%). Management projects AI semiconductor revenue alone will exceed $100 billion in fiscal 2027, with guidance implying 200%-plus growth in the current quarter.

Key customer contracts with Alphabet (through 2031) and Meta (through 2029) underpin the chip-side visibility. The software side adds a second compounding layer—enterprise VMware upgrades responding to AI-era security demands—giving AVGO a dual-engine growth profile that pure-play chip vendors cannot easily replicate. The stock currently trades roughly 20% below its May 2026 peak, a gap that reflects concentration risk from two dominant hyperscaler customers rather than any fundamental deterioration in the business.

Broadcom (AVGO) ended the session at $427.49, up 1.65%, outpacing both the S&P 500 (+0.62%) and the Nasdaq (+1.30%). The chipmaker also outperformed its Computer and Technology sector peers over the past month, posting a +4.85% gain against a sector that slipped 0.09% in the same period.

Valuation metrics suggest the move has fundamental backing. Broadcom trades at a forward P/E of 35.83, a discount to its industry average of 38.41, and carries a PEG ratio of 0.7 versus the sector's 1.15 — indicating growth expectations are not yet fully priced in. Analysts project next-quarter EPS of $3.22, up 90.53% year-over-year, on revenue of $29.47 billion, up 84.74%. Full-year consensus sits at $11.74 EPS and $106.07 billion in revenue. The stock holds a Zacks Rank of #2 (Buy) and ranks in the top 21% of tracked industries, with consensus EPS estimates revised 0.03% higher over the past 30 days.

Broadcom is increasingly framed as the preferred chip stock for investors seeking AI infrastructure exposure without the volatility of merchant GPU suppliers. Analysts point to its custom accelerator contracts with Google, Meta, OpenAI, and Anthropic—backed by $30 billion in AI bookings in Q2 FY2026 and contract visibility extending into 2028—as evidence of a "tollbooth" position on the AI buildout. CEO Hock Tan targets over $100 billion in fiscal 2027 AI revenue.

The investment case rests on a combination of locked-in demand and software-like margins: adjusted EBITDA of 69% and free cash flow of $10.26 billion in Q2. The stock trades at a forward P/E of 21 with a PEG ratio of 0.44, and carries 44 analyst buy ratings with zero sells, with a consensus price target of $527.88.

Broadcom reported Q2 FY2026 AI semiconductor revenue of $10.8 billion, up 143% year-over-year, with total revenue reaching $22.19 billion (+47.9%). The company guided Q3 AI revenue to $16 billion — a 200%-plus increase — and set a full-year AI semiconductor revenue target of $56 billion, with forward visibility extending to 2028. AI bookings exceed $30 billion against $10.8 billion already shipped. A recent analysis argues this performance positions Broadcom as a more structurally significant challenge to Nvidia than AMD, because rather than competing head-to-head in the GPU market, Broadcom designs proprietary XPUs for hyperscalers including Google, Meta, OpenAI, and Anthropic, and supplies Ethernet networking fabric (Tomahawk 6, Jericho 4) that interconnects large GPU clusters — with networking alone accounting for roughly 40% of Q2 AI revenue.

The distinction matters: AMD must win customers away from Nvidia, while Broadcom collects what the analysis calls a "toll" on hyperscaler infrastructure regardless of which compute vendor wins. Hyperscalers are actively engineering around dependence on Nvidia GPUs, and custom silicon contracts give Broadcom a role in that substitution rather than making it a casualty of it. AMD reported its own strong Q2 results — data center revenue doubled to $6.72 billion, total revenue reached $11.54 billion (+50.1%), and it secured a major Anthropic deal covering up to 2 gigawatts of MI450 capacity — but its upside remains tied to displacing Nvidia in merchant GPU sales, a structurally narrower position than Broadcom's dual custom-silicon and networking footprint.

Analysis framing a "bipolar silicon" market structure argues that hyperscalers deliberately maintain both Nvidia and Broadcom as AI chip suppliers to avoid vendor lock-in, with each occupying a structurally distinct role. Under this view, Nvidia controls the frontier training platform with its full merchant GPU stack, while Broadcom competes on inference economics through custom ASICs co-designed with hyperscale customers — optimizing cost-per-token at massive scale rather than selling a generalized product. Broadcom reported $10.80 billion in AI silicon revenue for Q2 FY26, up 143% year-over-year, on total quarterly revenue of $22.19 billion, with CEO Hock Tan guiding Q3 AI semiconductor revenue above $16.0 billion — more than 200% growth year-over-year.

Broadcom's custom ASIC approach gives it entrenched positions with a concentrated set of hyperscale co-designers, most visibly through a reported $30 billion Apple custom chip partnership running through 2031, as well as dominance in Ethernet AI networking switches. The company targets $100 billion in cumulative AI sales by 2027. The bipolar structure thesis suggests hyperscalers have an incentive to keep both suppliers viable, which insulates Broadcom from Nvidia's competitive gravity while differentiating its revenue profile toward inference workloads and recurring capital returns — the company has raised its dividend for 15 consecutive years.