![]() |
Last week, while going through some market news, I noticed AMD had just crossed the $10 billion revenue mark. A solid 38% jump year over year. Even now, looking at the chart hovering around $460, I keep thinking about the conversations I've had with colleagues. Whenever I bring up AMD, the response is almost always the same. "They probably just got lucky with the AI chip boom."
That's missing the point entirely.
When I dug deeper, I found something far more interesting than a revenue beat. AMD didn't just sell chips to OpenAI and Meta. They handed them warrants — the right to purchase AMD stock in the future at a price of just one cent per share. We're talking up to 160 million shares for OpenAI and another 160 million for Meta. 320 million shares total.
Most people hear that and move on. But sit with it for a second. Why would a semiconductor company give away equity instead of just selling chips? By the time you finish reading this, you'll understand exactly what Lisa Su pulled off. She didn't just acquire customers. She turned them into financial stakeholders who now have every reason to want AMD to succeed.
From PC Parts to AI Factory Core
AMD posted $10.25 billion in first quarter revenue, and this is a company moving with the energy of a growth stock again. But where that money is coming from matters more than the headline number.
Data center revenue hit $5.8 billion, up 57% year over year. That tells you everything about what AMD has become. This is no longer the company you think of for PC CPUs or gaming chips. The real engine now is the data center, packed with EPYC server processors and Instinct GPUs handling the heavy lifting of AI computation. AMD is transitioning from a computer parts vendor into a core infrastructure provider for AI factories. And it's working.
Exploiting the Fear of Big Tech
Nvidia sits at the center of the AI GPU market. And because of that, every major tech company is carrying the same quiet anxiety underneath their confidence.
Is it really smart to hand our entire AI infrastructure over to one supplier?
No matter how dominant a single vendor is, building your whole business on their supply chain is a risk that keeps executives up at night. AMD spotted that crack and stepped right into it.
The OpenAI deal involves a deployment at 6 gigawatt scale, starting with a 1GW rollout based on the MI450 chip in the second half of 2026. Meta isn't just buying GPUs either. They're pulling in a full package that includes EPYC CPUs, ROCm software, and the Helios rack architecture.
Think of it this way. If the GPU is the muscle generating raw power, the CPU is the nervous system deciding when and how to use it. Lisa Su isn't dropping off chips at the loading dock. She's embedding AMD into the anatomy of her customers' AI operations.
![]() |
| While GPU provides the raw strength, the CPU dictates the move |
Financial Engineering Hidden Inside a Chip Deal
The warrant structure is where this gets genuinely clever.
These warrants let the holder buy AMD stock later at a predetermined price. In this case, that price is one cent per share. On the surface it looks like free equity. But the conditions attached are anything but loose.
For OpenAI, the first tier of warrants only unlocks once the initial 1GW deployment is actually installed and running. Additional rights open up as their purchasing scale expands toward the full 6GW. And the final tier is tied to AMD's stock reaching $600. Meta has a similar structure, with warrants unlocking based on Instinct GPU shipment milestones.
"The moment these warrants are attached, OpenAI and Meta are no longer just customers. Their financial interests are pointing in exactly the same direction as AMD's."
The more AMD chips they deploy, the stronger AMD's earnings look. The stronger AMD's earnings, the higher the stock goes. And the higher the stock goes, the more those warrants are worth. Lisa Su embedded financial engineering directly into a semiconductor supply contract. That's not a standard vendor relationship. That's something else entirely.
Turning Weakness Into a Strategic Weapon
Let's be honest. AMD is not the dominant force in AI GPUs. Nvidia is. But instead of competing on price alone, Lisa Su went after the three things her biggest clients actually care about most: supply security, freedom from single-vendor dependency, and a tangible reward for diversifying.
OpenAI gets a second reliable source for its infrastructure. Meta gets a custom GPU and CPU roadmap built around their specific needs. And both of them now have a financial stake in AMD's stock performance.
Addressing the fear, the greed, and the cost calculations of the world's most powerful tech companies all at once, and getting them to align their interests with yours, is an extraordinary move. That's the real story here.
The Investor's Reality Check
As compelling as this all sounds, existing shareholders need to look at it with clear eyes.
320 million shares in warrants is not a small number. Stock dilution is a real consideration and it deserves serious attention.
The first thing to watch is whether that initial 1GW deployment for OpenAI and Meta actually ships on schedule. The market doesn't reward press releases. It rewards physical silicon running inside data centers and showing up as revenue on an earnings call.
Beyond that, the $5.8 billion in data center revenue needs to keep growing, and margins matter just as much as the top line. There's a real possibility that AMD offered significant incentives to land these clients. If that's the case, the revenue numbers might look impressive while the actual profitability tells a quieter story. Growth paired with weak margins is not the signal investors are looking for.
![]() |
| Navigating the fine line between explosive growth and stock dilution |
Finally, watch how the warrant tiers unlock over time. Each one is a dual signal. On the positive side, it means the customer is actively deploying AMD hardware at scale. On the other side, it brings incremental dilution. The real confirmation that this strategy is working will come when EPYC CPUs and Instinct GPUs are running together as an integrated system inside these data centers. When the nervous system and the muscle work as one, that's when AMD's case as a true AI infrastructure player becomes undeniable.
Rewriting the Rules of the AI War
The AI chip market is no longer a straightforward competition decided by benchmark scores and performance charts. The game has changed.
Lisa Su's approach is bold and carries real risk. But the logic behind it is sound.
"The moment these announcements become actual shipments and show up as data center revenue, AMD stops being an underdog story and starts being something much bigger. A new pillar in the AI infrastructure market."
Rather than reacting to daily stock movements, the smarter move is to follow the mechanics underneath the contracts and watch where the capital is actually flowing. That's where the real picture is. Hope this helps you see it more clearly.
Related Insight: [The Subscription Trap: Are you a customer, or a digital serf?]
Curious about the real stories behind big tech, crypto, and everyday economics? 👉👉👉Subscribe to The Techtonic for your weekly dose of easy-to-read business trends.




댓글 쓰기