Qualcomm stock surged on Tuesday after the chipmaker unveiled a sweeping multi-generation AI infrastructure partnership with Amazon, giving investors their clearest sign yet that Qualcomm’s push beyond smartphones is gaining traction with the world’s largest cloud companies. The agreement will see Qualcomm and Amazon collaborate on customized silicon for artificial-intelligence inference, while the companies will also develop high-speed optical connectivity products capable of supporting data rates of up to 1.6 terabits per second inside AI data centers. The headline number is even more striking: an SEC filing shows that purchases tied to the agreement can reach as much as $60 billion, with Amazon receiving warrants to acquire up to 25 million Qualcomm shares as commercial milestones are met.
That explains why Qualcomm stock jumped sharply after the announcement, with the shares climbing more than 8% at one point during Tuesday morning trading. The market is not simply celebrating another customer win. Investors are beginning to see a possible route for Qualcomm to transform itself from a company still heavily associated with smartphone processors and modems into a much broader supplier of AI infrastructure. Amazon joins Meta as another major hyperscaler validating that strategy, and the timing could hardly be more important as Apple continues reducing its reliance on Qualcomm modems and the traditional handset market offers nowhere near the growth available in AI data centers.
The deal does not mean Qualcomm has already booked $60 billion of revenue. That figure represents the maximum purchase threshold associated with the warrant structure, not a guaranteed contract value. But the agreement is still unusually significant because Amazon has already made enough initial purchase commitments for 3.75 million warrant shares to vest immediately. The rest will vest in stages as Amazon enters additional commercial agreements, places binding orders and makes actual purchases. For investors trying to decide whether Qualcomm’s ambitious data-center targets are realistic, that structure gives the partnership far more weight than a vague technology memorandum.
The $60 Billion Number Comes With an Important Catch
Qualcomm’s September 8 SEC filing reveals the financial architecture behind the deal. Amazon received a warrant allowing it to purchase up to 25 million Qualcomm shares at an exercise price of $161.26 per share, potentially representing an investment of just over $4 billion if the full warrant is exercised at that price. The warrant expires in September 2036, providing a decade-long window during which the commercial relationship could expand.
Crucially, however, the warrant shares do not vest simply because Amazon signed the partnership announcement. They are tied to execution. Qualcomm said vesting depends on commercial arrangements, binding purchase orders and actual spending on server-chip products, systems, technologies and manufacturing services, with the final tranche connected to cumulative payments of up to $60 billion. Reuters described the arrangement as giving Amazon the ability to purchase as much as $60 billion of Qualcomm AI data-center products over time.
Related articles
That makes the transaction simultaneously more exciting and more complicated than the headline suggests. Qualcomm has not suddenly secured $60 billion of guaranteed revenue, but Amazon has been given a powerful financial incentive to deepen the relationship as it buys more products. If Qualcomm executes successfully and Amazon’s spending moves toward the upper end of the structure, the resulting revenue could dramatically reshape a data-center business that is still small today.
The market appears to understand that distinction. Investors reacted aggressively because the deal offers evidence that Qualcomm’s data-center roadmap is attracting real hyperscaler interest, not because $60 billion can immediately be inserted into an earnings model.
Amazon Is Buying More Than Just AI Chips
The partnership is also broader than a simple processor order. Qualcomm and Amazon plan to work together across multiple generations of custom silicon designed for AI inference, the stage where trained models respond to users, operate software agents and process production workloads. That distinction matters because inference could ultimately represent one of the largest recurring sources of AI compute demand as companies move from training models to deploying them across millions of applications and users.
The companies are also developing optical connectivity technology reaching up to 1.6T speeds. AI data centers increasingly require enormous amounts of bandwidth because thousands of processors must exchange data rapidly enough to behave like one massive computing system. As cluster sizes increase, networking and optical connectivity become just as important as the processors themselves. Qualcomm’s expertise in SerDes technology, signal processing and low-power silicon gives it an opportunity to sell more than one component into the same infrastructure buildout.
Qualcomm will deepen its own use of AWS at the same time, including Amazon Bedrock for electronic design automation workloads. The objective is to use AI to shorten chip-development cycles, turning the partnership into a two-way relationship in which Amazon buys Qualcomm silicon while Qualcomm consumes more Amazon cloud infrastructure.
This broader architecture helps explain why Qualcomm believes the data-center opportunity can become large enough to matter at the corporate level.
Targeting More Than $15 Billion in Data-Center Revenue
Only a few months ago, Qualcomm laid out a remarkably aggressive target: more than $15 billion in annual data-center revenue by fiscal 2029. At the same June investor event, management doubled its broader fiscal 2029 non-handset revenue target to $40 billion and said it expects non-GAAP EPS to exceed $18.
Those goals initially looked ambitious because Qualcomm was entering a market where Nvidia dominates AI accelerators, Broadcom has entrenched custom-silicon relationships and AMD is fighting aggressively for server share. Qualcomm also acknowledged that it was late. Yet management argued that decades of designing high-performance, power-efficient chips for mobile devices provide exactly the expertise increasingly required by energy-constrained AI data centers.
The customer list now gives that argument more credibility. Meta has already selected Qualcomm’s Dragonfly C1000 CPU for future server deployments beginning in the second half of 2028, making Meta the first publicly named major customer for Qualcomm’s new data-center CPU strategy. Qualcomm said at the time that it had also secured additional hyperscale engagements. Amazon now gives investors another named customer—and one with a potentially enormous commercial framework.
That does not guarantee Qualcomm reaches $15 billion by 2029. But a target that once depended heavily on management projections now has two major hyperscaler relationships sitting underneath it.
Amazon’s AI Spending Makes This Customer Especially Valuable
Amazon is not a normal semiconductor buyer. It is one of the largest infrastructure spenders on earth, and its appetite for AI capacity is still accelerating. Amazon raised its 2026 capital expenditure plan to roughly $220 billion after second-quarter AWS revenue jumped 37%, its fastest cloud growth in more than four years.
That spending creates an enormous addressable market for suppliers capable of lowering the cost of AI infrastructure. AWS already designs its own Trainium AI accelerators, Graviton CPUs and Nitro chips, meaning Amazon does not need Qualcomm simply to replicate off-the-shelf silicon. The fact that it is partnering with Qualcomm suggests the opportunity lies in specialized custom components where Qualcomm’s low-power design expertise and connectivity intellectual property can add something AWS does not want to build alone.
Reuters reported that Amazon’s custom-chip operation is already running at approximately a $25 billion annualized revenue rate, highlighting just how large the company’s internal silicon strategy has become.
For Qualcomm, gaining access to even a portion of that infrastructure ecosystem can matter enormously. Amazon is also spending at a scale where successful components can quickly move from pilot programs into very large deployments, especially when they help improve power efficiency or reduce the cost of inference.
Power Efficiency Could Be Best Weapon Against Bigger Rivals
Qualcomm is not going to beat Nvidia by copying Nvidia. Its strongest argument comes from a different part of its history: building extremely power-efficient processors for battery-powered devices.
That skill set has become unexpectedly valuable in data centers because electricity is now one of the biggest constraints facing the AI industry. Hyperscalers are spending hundreds of billions of dollars on new infrastructure, yet many projects are being delayed by grid connections, data-center construction and the difficulty of supplying enough power to massive clusters. Qualcomm believes that chips delivering more computation per watt can therefore become strategically important even in facilities where batteries are irrelevant.
Inference makes that argument particularly interesting. Training the largest frontier models requires extreme raw performance, an area where Nvidia’s ecosystem remains exceptionally strong. Inference can prioritize a different combination of price, latency, energy efficiency and workload specialization, potentially creating more room for custom silicon.
The Amazon agreement therefore does not need to turn Qualcomm into the next Nvidia to become financially transformative. If Qualcomm captures a meaningful niche in custom inference chips, CPUs and optical connectivity, management’s $15 billion data-center target could become achievable without taking dominant market share.
Qualcomm Badly Needs a New Growth Engine
The bullish reaction also reflects what Qualcomm is trying to escape. The company remains exposed to a mature smartphone market, while Apple is progressively shifting more devices toward internally developed modem technology. Those pressures have made diversification one of the central questions surrounding QCOM stock.
Qualcomm has already pushed aggressively into automotive, PCs, industrial IoT and robotics, but data centers offer a fundamentally larger opportunity because of the extraordinary capital flowing into AI infrastructure. Management now expects non-handset businesses to generate $40 billion of revenue by fiscal 2029, roughly twice its previous target.
The Amazon partnership therefore arrives at exactly the moment investors need evidence that diversification is becoming more than a presentation slide. Meta gave Qualcomm its first major data-center validation. Amazon makes the case much harder to dismiss.
Still, the timing also creates execution risk. Qualcomm must develop products that meet Amazon’s cost and performance requirements, bring them into volume production, and compete against companies that already have years of data-center experience. A decade-long warrant does not remove those challenges.
The $60 Billion Headline Is Huge, but Execution Is the Real Catalyst
The immediate excitement around the stock is understandable. Amazon can potentially purchase up to $60 billion of Qualcomm server-chip products, systems and related services over the life of the arrangement, while the warrant structure gives Amazon a direct economic incentive to deepen the partnership. Qualcomm is simultaneously gaining another marquee hyperscaler customer just months after announcing a goal of more than $15 billion in annual data-center revenue by 2029.
But investors should not confuse potential with booked revenue. The $60 billion threshold is tied to future purchases and warrant vesting; it is not a guaranteed backlog. The real milestones will come when Qualcomm begins disclosing meaningful data-center revenue, additional production deployments and evidence that its products can compete economically against established AI infrastructure suppliers.
That is what makes Tuesday’s announcement so important. Qualcomm did not merely announce another AI initiative. It gave investors a credible mechanism through which one of the world’s largest cloud operators could become an enormous customer.
If Amazon purchases ultimately approach even a meaningful fraction of the agreement’s maximum, Qualcomm’s identity as a smartphone-chip company may start looking increasingly outdated.
The $60 billion number grabbed Wall Street’s attention. Now the company has to turn that possibility into silicon, revenue and earnings.
Disclaimer
This article is for informational purposes only and does not constitute financial or investment advice. Readers should conduct their own research and consider consulting a qualified financial advisor before making investment decisions. This article was researched and drafted with the support of AI, then reviewed, fact-checked and edited by the editorial team before publication.
