Is AMD Stock a Good Investment? The Bull and Bear Case

When a stock multiplies in value within a year, the people who own it and the people who missed it ask the same question: is it still a good investment? The usual places to look give you a letter grade, an analyst's price target or a thread of strangers arguing. Each tells you what someone concluded, but it hides the reasoning you would need to judge it for yourself.

A price chart cannot settle it either. A rising price shows that buyers were willing to pay more, and it says nothing about whether the business will earn enough to justify what they paid. Two people can study the same chart and reach opposite conclusions, because each of them is betting on a different future.

A better way to answer is to set out both cases with their numbers. What has to go right for the stock to be worth its price, what could go wrong, and how much good news does the price already assume? Then the question becomes which case you find more convincing, and whether you could keep holding while the market changes its mind.

Advanced Micro Devices, the maker of EPYC server processors and Instinct AI accelerators, is an unusually clear case to work through. Its management has set large public targets, and its own filings list the risks in plain words.

Is AMD stock a good investment?

AMD stock is a good investment at its September 25, 2026 price only if AMD meets management's target: adjusted earnings per share about three times their current pace within three to five years of November 2025. That price was about 161 times past-year earnings. Bulls point to AI demand and server share, bears to the price, Nvidia and execution.

Which case wins depends on numbers AMD has not reported yet, above all its next few quarters of Data Center revenue and adjusted earnings. A share of AMD stock is a claim on the profits of a company whose price already assumes several years of fast growth.

The ratios, the market value, the price change and the beta are Stock Analysis figures at the September 25, 2026 close (source: Stock Analysis, AMD statistics). A P/E, or price-to-earnings ratio, is the quickest way to see how much hope a price holds. It tells you how many dollars buyers pay for each dollar of yearly profit per share.

The bull case for AMD stock

The bull case says AMD is becoming one of the main suppliers of chips for AI data centers. It expects profits to grow fast enough to make today's price look reasonable within a few years. AMD's second-quarter 2026 results, its raised targets and three large AI agreements are the evidence bulls point to.

In the quarter to June 27, 2026, AMD reported revenue of $11.5 billion. Its Data Center segment, which sells server processors and AI accelerators, brought in $6.7 billion, up 107% from a year earlier (source: AMD second-quarter 2026 results, August 4, 2026). For the third quarter, AMD's guidance, its own forecast for the quarter, was about $13.0 billion of revenue, plus or minus $300 million. That would be roughly 13% more than the second quarter, and AMD guided its adjusted gross margin, the share of sales left after production costs, to about 56%, unchanged from the second quarter.

Management said it expects to beat its long-term targets. On the August 4 earnings call, chief executive Lisa Su said AMD now expects revenue to grow substantially faster than its earlier target of more than 35% a year. She also said AMD expects to significantly exceed its $20 annual earnings-per-share target within its strategic timeframe (source: AMD Q2 2026 earnings call transcript). AMD set that target at its Financial Analyst Day in November 2025 as non-GAAP earnings per share above $20, among growth targets for the next three to five years (source: AMD, November 11, 2025). Non-GAAP, or adjusted, earnings depart from generally accepted accounting principles (GAAP) by leaving out costs such as stock-based pay and the amortization of intangible assets bought with past acquisitions. AMD's adjusted earnings per share were $1.66 in the second quarter. Four quarters at that pace would come to about $6.64, so the target asks for roughly three times as much.

Three large agreements back those targets. Under multi-year agreements signed in October 2025 and February 2026, OpenAI and Meta each plan to deploy up to 6 gigawatts of AMD's data-center graphics processing units, or GPUs (source: AMD Form 10-Q, Q2 2026). Anthropic followed in July 2026 with up to 2 gigawatts of MI450 Series GPUs, starting with a first gigawatt in the first half of 2027 (source: AMD and Anthropic, July 22, 2026). A gigawatt here measures the electrical power of the data-center capacity being built, and Su has put the revenue from each gigawatt in the double-digit billions of dollars. Each deal sets a ceiling on what that customer may buy, and turns into revenue only as the customer deploys.

Server processors add a second growth source. Mercury Research estimates that AMD had 34.5% of server processors in the second quarter of 2026, up from 27.3% a year earlier, while server processor shipments across the market rose 20% over the same year (source: The Register, August 21, 2026). AMD's quarterly filing credits Data Center growth to both EPYC processors and Instinct MI350 Series GPUs, so the build-out lifts both product lines.

On the same call, Su said Helios, AMD's rack-scale AI system, meaning a full rack of servers built to work as one machine, was in production. Initial shipments were due to begin later in the third quarter, which ended on September 26, 2026. AMD had not said by September 28 whether they began, and its third-quarter report will show it. Those shipments are the next step in the AI and data-center story behind the bull case.

The bear case for AMD stock

The bear case accepts most of the good news but argues that the price has run ahead of it. At about 161 times past earnings and 25 times past sales, AMD needs years of growth to arrive on time. Meanwhile, the risks that could slow it down are listed in AMD's own filings.

Nvidia is the main competitive threat. AMD's quarterly filing says Nvidia uses its data-center GPU position, its money and its "proprietary software ecosystem" to promote its systems and influence AMD's customers, and that this may limit AMD's market share and margins (source: AMD Form 10-Q, Q2 2026). The same passage names a lack of software interoperability among the practices that can limit customers' ability to choose AMD. Interoperability is whether software written for one company's chips runs on another's, so a customer that switches has to be confident its own code will run as well on AMD's.

Execution is the next risk. In February 2026, the research firm SemiAnalysis reported that manufacturing delays would push mass production of AMD's first MI455X rack system to the second quarter of 2027. An AMD vice president replied publicly that the system was on target for the second half of 2026 (source: Tom's Hardware via Yahoo Finance, February 17, 2026). Su's August statement is the company's latest word, and AMD's next quarterly report will show whether initial Helios shipments began in the third quarter. The same 10-Q warns that more complex new products may bring quality or production issues and product delays, and the MI450 and Helios roadmap sets the dates to check against.

Dilution is the last argument. Dilution means more shares, so each existing share owns a smaller slice of the company. AMD issued warrants to OpenAI and Meta for up to 160 million shares each, and a warrant here is a right to buy shares at $0.01 each. The shares vest in stages as each customer buys more Instinct chips and as AMD's stock price reaches set targets, and none had vested by June 27, 2026 (source: AMD Form 10-Q, Q2 2026).

Say you own 1,000 AMD shares. If every tranche of both warrants vested and was exercised, up to 320 million new shares would join the 1.63 billion outstanding on July 29, 2026. That is about 19.6% more shares, and your 1,000 shares would then own about 16% less of the company than before. That dilution happens only if AMD sells the chips and its share price hits the targets. Bulls call it the price of success, and bears call it a cost the headline figures leave out.

How much growth is already in AMD's price?

The price-to-sales ratio is the plainest way to see how much growth the market already expects, because sales are harder to flatter than profits. At the end of fiscal 2025, AMD's market value was about 10 times its sales for that year. By September 2026, it was about 25 times its sales over the past 12 months.

Put another way, buyers were paying about $25 for each dollar of yearly sales, against about $10 nine months earlier (source: Stock Analysis, AMD financial ratios). Sales did grow, from $34.6 billion in fiscal 2025 to $41.3 billion over the 12 months to June 2026, a rise of about 19%. Over roughly the same stretch, AMD's market value rose from about $350.0 billion to about $1.0 trillion, a rise of about 194%. The gap between those two numbers is good news the market has already paid for, and a buyer at today's price is counting on the next few years to deliver it.

Say you buy one share at $630.63. At a trailing P/E of 160.96, you pay about $161 for each dollar of earnings per share that AMD made over the past year. If those earnings stayed flat, it would take about 161 years of them to add up to your purchase price. Nobody buys AMD expecting flat earnings, of course. The bull case needs earnings to grow several times over. Management's target is set in adjusted earnings, which run above the reported earnings behind this P/E, but on either measure the P/E falls quickly if that growth arrives.

A high P/E is not new for AMD, and on its own it proves little. At the end of fiscal 2023, the trailing P/E stood at 278.85, higher than today, because profit was small that year. AMD's net income for 2023 was $854 million, about a fifth of the $4.3 billion it earned in 2025 (source: Stock Analysis, AMD financials). A P/E can be high because the price is rich or because profit is temporarily small, and the ways AMD stock is valued differ in which of those they catch.

AMD stock drawdowns in 2026

A drawdown is a fall from a previous high. How deep AMD's drawdowns run matters as much as how high it climbs, because an investor who sells during a fall never collects the recovery. Three of AMD's sharp falls between its June high and September record show the pattern, and only one followed a change in its own numbers.

AMD's price history shows the June-to-July slide taking 26% off the stock in four weeks (source: Stock Analysis, AMD price history). The Motley Fool tied it to the DeepSeek report and the worry about AI spending (source: The Motley Fool, August 10, 2026). In the essay behind the September 14 fall, Dario Amodei argued that "[w]e must slow the pace at which we improve the capabilities of AI models" (source: Dario Amodei, We Must Pace the Frontier). Chip stocks fell across the sector that morning (source: 24/7 Wall St, September 14, 2026). So two of those three falls came from news about the AI industry as a whole. Neither changed AMD's revenue, margins or outlook. The forces that move AMD's price on a given day include many that have nothing to do with AMD.

That matters because a bull case can be right and still be painful to hold. Say you bought at the June 30 close and sold near the late-July low. You would have lost about a quarter of your money on a thesis that, by September, had carried the stock to a new all-time high.

AMD stock risks and where each would show first

AMD stock's main risks are its valuation, Nvidia's lead in AI chips and delays to new products. The others are its reliance on Taiwan Semiconductor Manufacturing Company (TSMC), US export rules on China, a few large customers, dilution from warrants and swings in AI-spending sentiment. Each would show first in a known place, from AMD's own reports to sector news.

Most of these risks are set out in AMD's quarterly filing for the period to June 27, 2026 (source: AMD Form 10-Q, Q2 2026). The China figure comes from its annual report for fiscal 2025 (source: AMD Form 10-K, fiscal 2025).

Two of these risks deserve a closer look before you buy. AMD's dependence on TSMC means a problem at one supplier in Taiwan reaches every advanced AMD chip at once. And who buys AMD's AI chips decides how much one customer's change of plan would cost.

Which investor each case fits

The bull case fits an investor with a long horizon, because management's targets cover three to five years from November 2025. It also fits someone who could watch the stock fall by a quarter in a month without selling, and who holds AMD as one part of a wider portfolio.

The bear case fits an investor who thinks 24 times sales already pays for the good outcome, or who would have to sell during a fall like July's. It also fits anyone who wants income. AMD pays no dividend, and its annual report says it does not expect to pay one in the near future (source: AMD Form 10-K, fiscal 2025). Every dollar of return has to come from the share price. Finally, it fits a portfolio that already holds other AI chip stocks, because the July and September falls hit the whole sector on the same days.

Before you buy, write down three things. First, the number that would make you sell because the thesis broke, such as a quarter of falling Data Center revenue or a confirmed Helios delay. Second, the largest fall you could sit through without selling, measured against the 26% AMD lost in four weeks in 2026. Third, how much of your portfolio AMD would be, given a five-year beta of 2.48. If you can answer all three, you have a plan, whichever case you pick.

Comparing AMD with Intel, the other large maker of PC and server processors, can sharpen the choice, because AMD and Intel stock are bets on very different conditions.

Taking a view on AMD through BloFin

Some views on AMD last weeks, not years: you might expect the next results to disappoint. BloFin lists the AMDUSDT Perpetual, a contract that tracks AMD's share price and settles in USDT, and on September 28, 2026 it offered up to 20x leverage. You can go long on the bull case or short on the bear case.

A perpetual is a futures contract with no expiry date, and holding one differs from owning the stock. You own no shares and get no vote. Every eight hours, a funding payment can pass between long and short positions, so a position held for months carries a running cost or credit that a share never does. The AMD perpetual's price index and funding decide what a position earns or costs from day to day. Over a long horizon, those payments add up. Together with the risk of liquidation, the forced closing of a position whose margin runs out, they are the main differences between holding the perpetual and owning shares.

Say you open a long position worth 1,000 USDT at 5x leverage, so you post 200 USDT of margin. A 20% fall in AMD's price would cost the position about 200 USDT, which is all of your margin. The 26% fall from June 30 to July 29 was deeper than that, so the position would have been liquidated before the July low. Keeping leverage low enough to survive a fall of that size is what keeps the position yours through a month like July.

Looking to take a view on AMD through the AMDUSDT Perpetual? To get started, you'll need to first create a BloFin account, fund your account with cryptocurrency, and navigate to the AMDUSDT Perpetual trading page.

Frequently asked questions

Is it worth buying AMD stock now?

That depends on the price you pay as much as on the company. Stock Analysis records a close of $214.16 at the end of 2025 and $630.63 on September 25, 2026, a rise of about 194% in under nine months. A buyer now pays for growth that a buyer at the end of 2025 could have had for about a third of the price. The test is whether the bull case still convinces you at $630.63, and whether you could hold through a fall like July's. If either answer is no, the price asks for more confidence than you have.

Is AMD stock overvalued?

AMD is expensive against its own recent history, and whether that makes it overvalued depends on how fast earnings grow. Stock Analysis puts AMD's forward P/E, the price against estimated earnings over the next year, at 56.99 in September 2026. That compares with 38.76 at the end of fiscal 2025 and 46.64 at the end of fiscal 2021. Buyers are paying more for each dollar of expected profit than at either of those year-ends, so the growth has to arrive for the price to hold.

Will AMD hit $1,000 a share?

Nobody can know, but the arithmetic shows what it would take. AMD had 1,632,475,042 shares outstanding on July 29, 2026, so a price of $1,000 would make the company worth about $1.6 trillion. That compares with about $1.0 trillion at the September 25, 2026 close, so it would take a rise of almost 60%. Getting there needs earnings to grow into a larger market value or buyers to accept an even higher multiple. A stock split would change the per-share arithmetic without changing what the company is worth.

Where will AMD be in 5 years?

Five years is long enough for either case to play out, and a few dated markers fall inside that span. AMD's quarterly filing says OpenAI's warrant can be exercised until October 5, 2030 and Meta's until February 23, 2031, so the dilution question will be settled by then. On the August 4, 2026 earnings call, Lisa Su also said AMD expects the data-center AI accelerator market to reach about $1.4 trillion by 2030. Where the share price ends up depends on how much of that market AMD wins, and no one can put a number on that today.

Is AMD a better buy than Nvidia?

The two stocks are bets on different positions in the same market. In its annual report, AMD says that in data-center chips it competes primarily against Intel and Nvidia (source: AMD Form 10-K, fiscal 2025). So buying AMD is a bet that its Instinct GPUs keep winning AI customers, while buying Nvidia is a bet that Nvidia keeps them. Each has its own valuation, customers and risks, so Nvidia and AMD stock are best judged side by side.

How volatile is AMD stock?

AMD's price history on Stock Analysis shows how volatile the stock has been in 2026. It rose 13.00% in a single session on July 30 and 9.95% on September 21, and it fell 8.15% on July 28, all on days without an AMD results release. For a new investor, that volatility is the practical risk to plan for, because it limits how much of a portfolio AMD can sensibly be.

Researched and written by the BloFin Academy editorial team with AI-assisted drafting. All facts independently verified. Primary sources include AMD's second-quarter 2026 results release and earnings call of August 4, 2026, its Form 10-Q for the quarter ended June 27, 2026 and its Form 10-K for fiscal 2025 filed with the SEC, AMD's announcement of its Anthropic partnership of July 22, 2026, valuation, financial and price data from Stock Analysis at the September 25, 2026 close, Mercury Research data reported by The Register on August 21, 2026, AMD's Financial Analyst Day release of November 11, 2025, Dario Amodei's essay of September 2026, reporting by Tom's Hardware of February 17, 2026, The Motley Fool of August 10, 2026 and 24/7 Wall St of September 14, 2026, and BloFin's AMDUSDT contract and copy-trading data read on September 28, 2026, current as of September 28, 2026.

Nothing in this article constitutes financial advice, and nothing in it is a recommendation to buy, sell or hold AMD in any form. The bull and bear cases here are arguments built on dated figures, and neither is a prediction: AMD's growth depends on customer orders and product schedules that can change, and a price near 161 times past earnings can fall a long way if that growth arrives late. Whichever case you favor, trading it through a leveraged AMDUSDT Perpetual position risks liquidation if AMD's price goes the other way. Past performance does not indicate future results. Do your own research and consider your risk tolerance before you trade on BloFin.