Imagine a floor in a dark room.

Looks perfectly clean, right?

Now shine a special UV light on it.

Suddenly, footprints glow everywhere.

The same thing happens in the stock market.

Most investors stare at a chart and see nothing.

Just noise. Random price moves. Chaos.

But one group of investors knows how to shine a special light on those same charts.

And what they see changes everything.

They can spot gains like 104% in under 3 months …

387% in about 2 years …

And 756% in a single year.

Before anyone else has a clue.

What are they seeing that you're not? Click here to discover it for yourself.

To your new vision,

Chris Graebe
Editor, Weiss Ratings

11780 US Highway 1, Palm Beach Gardens, FL 33408-3080 Would you like to edit your e-mail notification preferences or unsubscribe from our mailing list?

Qualcomm Just Signed a $60 Billion Amazon Deal. The Structure of It Is More Interesting Than the Size.

Written by Evan Brooks · September 9, 2026

Start Here

Qualcomm (QCOM) announced a multi-generation custom AI chip partnership with Amazon Web Services on September 8. Amazon can purchase up to $60 billion in Qualcomm AI data center chips under the long-term agreement. QCOM shares rose more than 10% on the news.
The financial structure: Qualcomm is granting Amazon warrants for up to 25 million shares — valued at roughly $4 billion at the strike price of $161.26 — that vest proportionally with product purchases, per an SEC regulatory filing.
The deal targets AI inference workloads and includes optical interconnect development up to 1.6 Tbit/s. Qualcomm projects data center chip revenue of $15 billion by 2029. Amazon joins Microsoft and Meta as anchor hyperscaler customers for Qualcomm's data center push.

Why Amazon Is Paying $4 Billion for the Right to Buy More Chips

The warrant structure is the part of the Qualcomm-Amazon agreement that deserves more attention than it has received. Amazon is not simply purchasing chips — it is acquiring equity exposure to Qualcomm's data center trajectory at a fixed strike price of $161.26 per share, with warrants vesting as product purchases are made. That linkage between purchase volume and equity vesting is a financial alignment mechanism: the more chips Amazon buys, the more equity it earns, and the more its financial interest in Qualcomm's success grows. A similar structure appeared in Marvell Technology's custom chip deal with Alphabet's Google weeks earlier, when Google acquired the right to a stake worth up to $12.2 billion. The pattern — hyperscaler as customer and contingent equity holder simultaneously — is becoming a standard feature of large-scale AI chip partnerships, not a one-off financing arrangement.

For Qualcomm, the strategic logic is straightforward: the company built its reputation on mobile handset chips, a market where it faces declining revenue from the eventual loss of its Apple modem business and weaker consumer handset demand. Data center AI chips represent a growth runway that is structurally larger and longer than any mobile replacement cycle. The challenge has been convincing hyperscalers — who have their own in-house chip programs — that Qualcomm's inference-focused silicon is worth adding alongside Nvidia's dominant training processors. Amazon's Trainium3 has been shipping since early 2026 and is reportedly nearly fully reserved. AWS senior vice president Peter DeSantis confirmed separately that Amazon is in discussions to sell Trainium chips externally. That context reframes the deal: Amazon is adding a specialized inference partner, not replacing its internal program.

Iran bad → oil spikes → you pay more.

Iran deal → oil drops → you "get relief."

Six months later, rinse and repeat.

Think that's an accident?

The same banks advising the White House are trading oil options while the diplomats are still shaking hands.

One man who sat in THOSE rooms — who advised Saudi Arabia AND Kuwait — just went public with the method they use.

Get it before this offer disappears

Ad by Omnia Research

The Inference Market and Why Non-Nvidia Challengers Are All Targeting It

The Qualcomm-Amazon deal is focused explicitly on inference — the computation that runs when a trained AI model generates a response — rather than training, where Nvidia's H100 and H200 GPUs dominate. That market delineation is deliberate. Training runs are large, infrequent, and concentrated among a small number of frontier labs with established Nvidia relationships. Inference runs are continuous, high-volume, and cost-sensitive in ways that create an opening for power-efficient alternatives. Qualcomm's engineering position in mobile chips — where power efficiency is the primary design constraint — translates to inference workloads in a way it does not translate to training. The optical interconnect component of the deal, targeting speeds up to 1.6 Tbit/s, addresses the bandwidth bottleneck that becomes the binding constraint once the compute efficiency problem is solved.

The $15 Billion Target and the Conditions That Would Reach It

Qualcomm's stated target of $15 billion in data center chip revenue by 2029 requires the Amazon relationship to convert from a partnership announcement into sustained purchase volume at scale. Bob O'Donnell, chief analyst at TECHnalysis Research, described the deal to Reuters as exactly what Qualcomm needed to reassure the market that its data center ambitions were achievable — a statement that acknowledged the prior gap between those ambitions and their supporting evidence. The warrant structure provides that evidence in a form that is financially binding rather than simply contractual: Amazon's equity exposure grows only if purchase volume grows, which means the announcement carries different weight than a memorandum of understanding. The company already has Microsoft and Meta as data center customers. A third hyperscaler at the scale of AWS — whose annualized chip revenue across its own programs was running above $20 billion as of Andy Jassy's April 2026 shareholder letter — changes the revenue trajectory math for the 2029 target in a way that smaller customer additions would not.

Sources: Bloomberg · Reuters · QZ · RTE · Rolling Out · SEC regulatory filing · TECHnalysis Research

[email protected] is on TheRawCapital.com list because you opted in before the crowd caught on.

If we go quiet — look in promotions, updates, or wherever your inbox files things it hasn't figured out yet.

Unsubscribe. We'll part ways cleanly.

If something's off, you can reach us here

254 Chapman Rd Ste 208 Newark, Delaware 19702.

Privacy Policy