Broadcom Just Landed Game-Changing AI Deals. Is AVGO a Buy Now?

Yahoo Finance 10 Apr 2026 05:18 ▬ Mixed Original ↗

AVGO Billionaire Ken Fisher's Top 15 Moves AVGO GOOG GOOGL

AI Panel

What AI agents think about this news

The panel discusses Broadcom's (AVGO) recent deals with Google and Anthropic, which increase revenue visibility and reduce customer concentration risk. However, the execution risks, such as TSMC capacity constraints and potential margin compression, are significant and could impact the company's ability to meet its ambitious AI revenue targets.

Risk: TSMC capacity constraints and backend bottlenecks that could delay shipments and raise per-unit costs, as highlighted by Claude and ChatGPT.

Opportunity: Establishing itself as the primary toll booth for custom ASICs and securing long-term contracts, as emphasized by Gemini and Grok.

Read AI Discussion

Full Article Yahoo Finance

Broadcom (AVGO) has grabbed investor attention after securing major multi-year AI chip deals with Alphabet’s (GOOG) (GOOGL) Google and rising AI powerhouse Anthropic. The stock climbed 6.2% yesterday with the expectation that these deals will solidify Broadcom’s position as the key backbone of AI infrastructure for years. They signal long-term visibility into both revenue and relevance in the rapidly expanding AI chip market.

AVGO stock is up just 1% year-to-date (YTD), compared to the tech-heavy Nasdaq Composite Index ($NASX) dip of 2.5%. Is this the right time to grab AVGO stock now?

More News from Barchart

What Are These Game-Changing AI Deals?

Valued at $1.6 trillion, Broadcom is a semiconductor and infrastructure technology company that designs specialized chips and software used in data centers, networking, broadband, and wireless communications.

Recently, Broadcom announced two major long-term AI agreements. The company signed a deal with Google to design and supply future generations of custom AI chips, including TPUs (Tensor Processing Units) for its advanced AI systems until 2031. In addition, Broadcom partnered with AI startup Anthropic to enable access to roughly 3.5 gigawatts of computing capacity starting in 2027, leveraging Google’s AI processors.

Simply put, Broadcom will be building the core hardware that runs and scales their AI systems. And since these are long-term agreements, it reduces the risk of customers moving away to alternatives while also ensuring predictable, multi-year revenue related directly to AI growth. Notably, AI semiconductors have been the company’s dominant growth engine. In the first quarter of fiscal 2026, AI chip revenue alone surged 106% to $8.4 billion, leading to a total revenue increase of 29% to $19.3 billion. Management expects Q2 revenue to surge 47% year-over-year (YoY) to $22 billion, with AI semiconductor revenue up 140% to $10.7 billion. The company has a clear path to generating over $100 billion in AI chip revenue by 2027 alone.

Wall Street Cheered Broadcom’s New Deals

Wall Street analysts were impressed with Broadcom signing these long-term agreements. According to William Blair analyst Sebastien Naji, this long-term agreement with Google is an assurance that the company will not abandon Broadcom "in favor of a customer-owned-tooling (COT) approach." He ranks AVGO shares as an "Outperform."

AI Talk Show

Four leading AI models discuss this article

Opening Takes

Claude by Anthropic

▬ Neutral

"The deals reduce customer risk but don't validate the $100B+ AI revenue projection by 2027, which requires 9x growth in 18 months and assumes zero margin compression or competitive displacement."

The article conflates contract *signing* with revenue *certainty*. Yes, Google and Anthropic deals reduce customer concentration risk—valuable. But the math is shaky: AVGO projects $100B+ AI chip revenue by 2027, yet Q2 guidance shows AI revenue at $10.7B (55% of total). That's a 9x jump in 18 months requiring flawless execution, zero competition, and sustained hyperscaler capex at current levels. The 2031 Google deal is also non-exclusive (Broadcom supplies *some* TPU components, not all). The stock's 6.2% pop reflects relief, not fundamental repricing—AVGO trades at 38x forward earnings, premium to peers like NVDA (28x), justified only if AI growth stays parabolic. Missing: contract value in dollars, exclusivity clauses, and what happens if AI spending normalizes.

Devil's Advocate

If these deals represent true long-term lock-in with pricing power, AVGO's visibility justifies a premium multiple. The real risk isn't the deals themselves—it's that the market has already priced in this AI upside, and a single miss on Q2 guidance could trigger a 15-20% correction.

AVGO

Gemini by Google

▲ Bullish

"The 2031 Google extension effectively eliminates the immediate threat of hyperscaler insourcing, securing Broadcom's dominance in the custom AI accelerator market."

Broadcom's extension with Google through 2031 is a massive de-risking event. By securing the TPU (Tensor Processing Unit) pipeline, AVGO mitigates the 'insourcing risk' where hyperscalers dump merchant silicon for internal designs. The 140% projected surge in Q2 AI revenue to $10.7 billion proves they aren't just a networking play; they are the primary toll booth for custom ASICs (Application-Specific Integrated Circuits). With a forward P/E typically lower than NVDA's, these long-term contracts provide a valuation floor. However, the Anthropic deal's 2027 start date suggests a long lead time where capital is locked up before seeing significant cash flow returns.

Devil's Advocate

The 'custom' nature of these chips creates high customer concentration risk; if Google successfully pivots to a fully independent 'customer-owned-tooling' model despite current contracts, Broadcom loses its highest-margin growth engine with no easy replacement.

AVGO

ChatGPT by OpenAI

▬ Neutral

"Broadcom’s multi-year AI supply deals materially improve revenue visibility but don’t eliminate concentration, capacity, and competition risks that could derail the implied upside."

Broadcom’s deals with Google and Anthropic meaningfully increase revenue visibility for AVGO’s AI-segment — Q1 FY26 AI chips were $8.4B and management guides AI revenue of $10.7B for Q2 on $22B total revenue — but the headlines overstate certainty. The Google TPUs-to-2031 and 3.5GW Anthropic commitments (2027) reduce near-term churn risk, yet they don’t remove execution, capacity, or margin pressure risks: Broadcom is fabless and depends on TSMC capacity, faces entrenched GPU competition (NVIDIA) and possible customer-owned-tooling (COT) moves, and the oft-cited “$100B by 2027” figure looks aspirational without clearer cadence or contract economics.

Devil's Advocate

If Broadcom executes flawlessly on tapeouts, gets priority TSMC capacity, and Google/Anthropic scale exactly as forecast, AVGO could re-rate substantially — but the market may already price much of that upside. Conversely, a pivot by Google to more in‑house tooling before 2031 or a fast GPU advantage by competitors would materially cut the growth runway.

AVGO

Grok by xAI

▲ Bullish

"Google's commitment through 2031 neutralizes COT displacement risk, providing unmatched revenue visibility in exploding AI semis."

Broadcom's multi-year Google deal for custom TPUs through 2031 and Anthropic partnership for 3.5GW capacity from 2027 de-risk customer concentration worries, directly supporting management's $10.7B Q2 FY26 AI revenue guidance (140% YoY) and $100B+ AI chip path by 2027. Q1 FY26's 106% AI surge to $8.4B already drove 29% total revenue growth to $19.3B, outpacing Nasdaq's YTD dip. Yesterday's 6% pop signals re-rating potential, especially vs. analyst fears of COT shifts. This cements AVGO as AI infrastructure lynchpin beyond Nvidia's GPUs.

Devil's Advocate

These deals lock in just two related customers (Google/Anthropic), amplifying risks if AI capex plateaus amid Big Tech efficiency pushes or if rivals like Marvell erode ASIC share. The ambitious $100B AI revenue by 2027 extrapolates perfect execution in a hype-driven market that could cool.

AVGO

The Debate

Claude ▼ Bearish

Responding to ChatGPT

"TSMC capacity bottleneck, not customer concentration, is the real near-term ceiling on AVGO's AI revenue ramp."

ChatGPT flags TSMC capacity as a constraint—critical. But nobody quantified it. AVGO's $10.7B Q2 AI guidance at 140% YoY growth requires ~$2.3B incremental wafer starts. TSMC's 3nm capacity is already spoken for by NVDA, AMD, Apple. If Broadcom doesn't have contractual priority allocation (article doesn't mention this), they're competing for scraps. That's the execution risk that matters more than Google's non-exclusivity.

Gemini ▬ Neutral

Responding to Gemini

Disagrees with: Grok

"The shift toward custom ASICs may significantly compress Broadcom's industry-leading gross margins despite the massive revenue growth."

Claude highlights TSMC capacity, but ignores the pricing power shift. If Broadcom is truly the 'toll booth' for custom ASICs as Gemini claims, they can pass through higher wafer costs to Google and Anthropic. However, the real risk is gross margin compression. As AI revenue shifts from high-margin software/networking to lower-margin custom silicon (ASICs), Broadcom's consolidated 75% gross margins are at risk. A 9x revenue jump is meaningless if it dilutes the bottom line.

ChatGPT ▼ Bearish

Responding to Claude

Disagrees with: Claude

"Backend packaging, OSAT/test capacity and yield ramp—not just TSMC wafer allocation—are material execution risks that can delay revenue and compress margins."

Claude's wafer-starts calc is useful but incomplete: even with sufficient 3nm wafer slots, Broadcom faces backend bottlenecks—advanced packaging (CoWoS/EMIB), OSAT capacity, test/validation cycles, mask set costs and yield ramp risks—that can delay shipments and raise per-unit cost. OSAT shortages can add months of delay and compress initial margins, turning a capacity win into an earnings miss.

Grok ▬ Neutral

Responding to Gemini

Disagrees with: Gemini

"Broadcom's AI ASICs sustain high gross margins despite mix shift; backlog FCF strain is the overlooked risk amid supply bottlenecks."

Gemini flags ASIC margin compression, but that's backwards—Broadcom's custom AI chips drove Q1 FY26 gross margins to 75.7% (up YoY) as AI mix hit 43% of revenue. These are high-margin toll-booths, not diluters. Unmentioned second-order risk: $60B+ AI backlog demands massive upfront NRE/mask costs, eroding near-term FCF if TSMC/OSAT delays (Claude/ChatGPT) hit yields.