BiyaPay观察:AI放量背后的利润账,AMD毛利率要降,股价为何仍反弹?
In AMD's earnings report, there is a set of figures that appear contradictory but are all the more worth investors' scrutiny: gross margin is expected to decline next quarter, yet operating income has room to keep growing; earnings improved markedly this quarter, yet free cash flow actually decreased.
If you explain the stock price solely with "results beat expectations," it is easy to miss what is changing at this company. AMD is moving deeper into the data center market, and revenue scale, product mix, and capital requirements are all adjusting. What investors need to judge is whether the company can raise overall returns during expansion—not merely sell a few more batches of chips.
BiyaPay market data shows that on November 3, US Eastern Time, AMD reported third-quarter 2026 results: revenue of $13.377 billion, up 45% year over year; adjusted earnings per share of $2.09, up 74% year over year. At the same time, fourth-quarter adjusted gross margin is expected to be about 55%, below the third quarter's 56%.
Does this 1 percentage point mean competitive pressure, or a structural change during the product ramp-up phase? The answer will directly affect how the market understands AMD's future profit space.
Data centers contributed more than the entire revenue increase, and AMD's growth center has shifted
In the third quarter, AMD's data center revenue reached $8.546 billion, up 97% year over year, accounting for about 64% of the company's total revenue. In the same period last year, that share was still around 47%.
What better reflects the change is the increment. Data center revenue increased by $4.205 billion year over year, while the company's overall revenue increased by only $4.131 billion. In other words, this segment contributed about 102% of the net revenue increase, while other businesses collectively offset it slightly.
This does not mean AMD has only one business left. Rather, it means the force supporting this round of growth has become highly concentrated. In judging the company's performance going forward, PC and game console demand still warrants attention, but data center delivery cadence, product mix, and customer deployment will carry greater weight.
Consumer-side performance needs to be viewed separately. Client revenue grew 7% year over year and did not decline year over year; gaming revenue fell 40%, dragging combined client and gaming revenue down 8%. The decline of the combined segment should not be written as all consumer businesses contracting.
Data centers are also not a uniform profit pool. Server CPUs and AI accelerators face different demand, supporting costs, and product ramp-up processes. If certain products scale faster, that can both push up revenue and change the company's average gross margin.
Therefore, a rising data center share should not be directly understood as gross margin rising every quarter. What really matters is whether, after different products grow, the total gross profit left behind and operating income have increased.
Gross margin is 1 percentage point lower, so why could operating income still be 40% higher?
Gross margin measures how much gross profit is left from each dollar of revenue; operating income also requires deducting operating expenses such as R&D, sales, and administrative costs. The two are related, but they do not always move in the same direction.
The midpoint of the company's fourth-quarter revenue guidance is $17 billion, adjusted gross margin about 55%, and adjusted operating expenses about $3.75 billion. Putting these three items together makes it possible to see what the market may be calculating.
Based on a static calculation using the guidance, gross profit is about $9.35 billion; after deducting operating expenses, adjusted operating income is about $5.6 billion, for an operating margin of about 32.9%. Compared with the third quarter's adjusted operating income of $3.923 billion and operating margin of about 29.3%, this corresponds to about 43% profit growth.
It should be emphasized that $5.6 billion is an estimate based on company guidance, not a separately issued operating income forecast by the company, still less realized performance. Its premise is that revenue, gross margin, and expenses are all close to the above assumptions.
But this calculation explains an important point: even if each dollar of revenue leaves slightly less gross profit, as long as revenue increases faster and R&D and administrative expenses do not increase at the same proportion, overall operating income can still expand significantly.
This is the expense dilution brought by scale. For investors, 55% is worth tracking, but it cannot alone be used to conclude that the company's profitability is deteriorating across the board. Conversely, if new revenue requires matching high proportional expense growth, profit elasticity will also be weaker than this calculation suggests.
As the stock price rebounds, which part of the improvement are investors buying?
In regular trading on November 3, AMD closed at $483.90, up 3.65%. The post-earnings reaction extended into the next day's premarket: as of 5:15 a.m. US Eastern Time on November 4, BiyaPay quoted $519.70, up 7.40% from the previous regular close. This is a premarket point-in-time quote and cannot be treated as the final gain for the day.
The stronger stock price at least shows that the market has not viewed the 55% gross margin guidance as bad news on its own. The profit space brought by revenue expansion has also entered investors' considerations. However, the stock price is also affected by prior positioning, valuation, and the market environment, and the entire gain cannot be attributed to a single metric.
More noteworthy is that the market is not trading only the third-quarter report card. According to reports, AMD management expects revenue growth of more than 80% in 2027, above analysts' previous expectation of about 77%. This means that even if the fourth-quarter non-GAAP gross margin guidance has slipped, investors are still focused on whether the AI data center business can continue to scale and whether future revenue growth can cover the pressure brought by declining margins. Compared with quarter-to-quarter gross margin changes, long-term growth expectations may be one of the important factors driving the stock price rebound.
For those tracking AMD, post-earnings action is also not just about "bullish or bearish." Liquidity and bid-ask spreads during premarket and regular trading hours, as well as the funds prepared in the account, all affect actual execution. Investors can monitor market trends in real time through BiyaPay. The platform uses USDT as its core funding entry point and connects US and Hong Kong stocks, cryptocurrencies, commodity futures, and cross-border remittances through a single account system. It supports exchange between fiat currencies and digital currencies and provides zero-commission trading services for real US and Hong Kong stocks. A centralized entry point helps reduce switching between different tools, making currency exchange, trading, and subsequent funding arrangements flow more smoothly.
Trading tools address how to participate. Judging whether this rebound can continue still requires returning to a harder part of the earnings report: after profit increased, why did cash not keep pace?
Profit is growing fast, but cash is first tied up in inventory and receivables
In the third quarter, AMD's operating cash flow from continuing operations was $1.202 billion. After deducting $648 million in capital expenditures, free cash flow was $554 million. Compared with $1.558 billion in the second quarter, free cash flow decreased significantly and was also below $1.530 billion in the same period last year.
Here one cannot directly conclude that "profits are not real." Profit and cash inherently have a timing gap. Companies purchasing ahead for future orders, or having recognized revenue but not yet received payment, can both affect cash performance in the quarter.
But nor should this capital occupation be ignored just because AI demand is strong.
At the end of the quarter, AMD's inventory rose to $11.463 billion, up about 35% quarter over quarter; accounts receivable rose to $9.209 billion, up about 26% quarter over quarter, both faster than the 16% quarter-over-quarter revenue growth. The company said the related investment was to support data center demand and that it plans to continue investing in inventory and capacity for 2027 growth.
This means growth requires fronting part of the capital. Stockpiling can help ensure delivery, but only if products are delivered smoothly and customers pay as agreed will the upfront investment gradually return to the company's books. Management's judgment of demand still needs subsequent turnover and collections to verify it.
Next, there is no need to mechanically require inventory to fall immediately. When a business is expanding rapidly, it is not unusual for the absolute amount of inventory to continue rising. What is more informative is whether, after revenue scales, the growth of inventory and receivables gradually matches the business scale, and whether operating cash flow can improve accordingly.
If these metrics begin to support one another, weaker cash flow is more likely a timing gap during expansion; if after several quarters book growth still requires more and more capital support, the market will need to reassess the cost of growth.
The more important AI revenue becomes, the more valuation depends on delivery quality
Data centers becoming the main growth source gives AMD greater revenue space and also makes the company more dependent on execution in this area. Strong demand is only the starting point; wafers, advanced packaging, memory, system integration, and customer data center deployment must all coordinate to generate revenue.
Investors therefore need to distinguish three things: customers are willing to buy, the company can deliver, and after delivery the company can earn reasonable returns. The first supports market space, the second determines which quarter revenue falls into, and only the third determines how much value growth can leave for shareholders.
The gross margin cut also needs to be placed in this framework. If the decline mainly comes from product mix changes while revenue growth, expense dilution, and collections support one another, the market can accept a certain degree of gross margin fluctuation. If the product ramp is accompanied by continued price cuts, rising costs, and uncontrolled expenses, then even if revenue still grows, the valuation logic will weaken.
For AMD, a single-quarter revenue record is still not enough to answer whether the stock is expensive. The future earnings on which valuation depends include assumptions such as product scaling, controlled expense growth, and continued customer deployment. Once one of these is realized more slowly than expected, earnings forecasts and the valuation multiple the market is willing to pay may be affected at the same time.
This is also why, after strong earnings, investors still need to watch for expectation gaps. A good company, good performance, and a suitable purchase price are interrelated but must be judged separately.
In the next earnings report, look at margins and cash flow together
The positive change in AMD's latest earnings report is that data center expansion has already brought higher operating income, rather than remaining only at the product and order level. Fourth-quarter guidance further demonstrates the possibility of scale diluting expenses, but free cash flow still reminds investors that growth currently requires considerable capital support.
The most valuable verification going forward is whether three sets of numbers can support one another: revenue and total gross profit continue to expand, operating expense growth is lower than revenue growth, and operating cash flow gradually catches up with earnings.
If only revenue is delivered while margins and collections do not improve, the market may still question the quality of expansion; if gross margin fluctuates slightly but operating income and cash recovery continue to strengthen, looking only at 55% may instead underestimate the business change.
For this earnings report, what I focus on more is not which percentage point gross margin lost, but whether AMD can convert a larger business scale into higher shareholder returns. The premarket rebound provided the first round of price feedback; what truly determines subsequent valuation is how much of these profits ultimately turns into cash.


