AMD值得买入，但信心不高；当前价格456美元为公允价值，但更希望其下跌15%。

AMD is focused on expanding its Helios strategic data center solutions, and expects this to start in the fourth quarter and accelerate in the first quarter of 2028.

Therefore, AMD is well-positioned to benefit greatly from bringing this product to market. But does this make AMD stock a buying opportunity alongside the company's large share of the server CPU market?

AMD's stock has seen a slight downward trend since it reached its peak. Of course, ever since AMD's stock broke the $500 mark per share , I've been warning investors that it's too expensive, and that there are better options on the market.

Of course, I understand that AMD is a very popular stock , and I received a lot of negative feedback when I made those videos suggesting that AMD stock was overvalued when it exceeded $550 and approached $600 per share .

Looking back today, it turns out that avoiding AMD stock at those levels above $500 and $550 was the right decision, as it is now trading at $455 per share.

The last time I updated my rating for AMD was on August 12, 2026, when I told investors again that it did not look like a buying opportunity at those levels.

On August 12, AMD stock was trading at around $520 per share, and has since lost about $70 of its market value. Has its value finally become sufficient to justify its upgrade to the "buy" rating?

Well, one of the things I really appreciated about AMD was its ability to innovate and gain market share in categories dominated by strong companies.

It has done a strong job as the second player in critical categories while continuing to innovate. In addition, I believe that Lisa Su, the CEO of AMD, is one of the best CEOs in the market.

Revenue has climbed to over $41.3 billion , and analysts at AMD and Wall Street analysts who follow the company indicate that its revenue is likely to rise over the next few years as interactive artificial intelligence proliferates and the importance of central processing units in data centers increases.

In fact, AMD expects the server CPU market to reach $200 billion, or $220 billion to be more precise, by 2030, and AMD aims to capture a 50% market share in that category, which means it could generate $110 billion from this sector alone by 2030.

Compare that to its current revenue over the past twelve months, which is only $41 billion.

So, you can see that the size of its future prospects is excellent. Profitability and cash flow are not yet as profitable as competitors like Nvidia. Its total cash flow from operations to sales was 24% over the past twelve months.

It is approaching its highest levels of the previous decade, but it is still only about half the levels at which Nvidia operates. Even when you look at operating profit margins, it still represents a small fraction of the levels at which Nvidia operates.

Of course, AMD's revenue is still just a small fraction of Nvidia's revenue. If AMD's revenue rises to the levels that management and Wall Street expect, it is reasonable to assume that its profit margins could approach levels similar to Nvidia's because it operates on a similar business model.

Therefore, given the similarities, it is reasonable to assume that AMD's profitability may approach that of Nvidia. Even if it doesn't reach or surpass Nvidia's level, I do n't think AMD will ever be number one in this category , but a strong second place could still yield excellent returns for investors.

So, now that the stock price has fallen by an approximate amount over the past few months, it is trading at a more reasonable valuation at a forward price-to-earnings ratio of 29.5 .

It has fallen from its previous peak in the forward price-to-earnings ratio of around 37 earlier this year. In 2024, the stock was trading at much higher valuations based on its forward price-to-earnings ratio, but that was due to its recovery from its cyclical low .

Remember that he suffered that great low point with the decline in consumer spending on electronics, and he had not yet delivered impactful products for the artificial intelligence sector in data centers.

Therefore, the forward price-to-earnings ratio appeared higher because the earnings per share were significantly lower . This is still the case to some extent.

Earnings per share today are much lower than they will be 3 or 5 years from now. The reason is that it is still recovering from those low levels. It is still in the early stages of capitalizing on the momentum of artificial intelligence. Therefore, today I also updated my fair value estimate for AMD. I now estimate the company's value to be $428 per share.

The market price is $456 per share . This falls within my margin of safety. Therefore, the stock appears to be fairly valued if we take into account the margin of safety , whether you measure that on a forward price-to-earnings basis or using a discounted cash flow model, I would say that AMD stock appears to be fairly valued .

I mentioned that I had n't rated AMD stock as a "buy " several months ago because I felt the valuation was overrated, and it turned out to be the right decision despite the pressure from many viewers and fans of AMD stock and the negative comments, as the stock was very popular and had a lot of media buzz as one of the most sought-after stocks in the market. Well, since the share price has dropped, it has become much more attractive .

It is fairly priced now. I don't think the stock is overvalued anymore . Therefore, given that it is one of the best-run and best-managed companies, with strong momentum supporting its growth, I am today upgrading AMD stock to "buy" at this lower price.

Previously, I would classify it as a "hold" or "not a buy" with a moderate level of conviction. My upgrade of the stock to a "buy" rating comes with low conviction and limited confidence.

I would feel more comfortable upgrading my rating on AMD stock or buying it at a market price about 15% lower. However, given the current situation, I list it as a buy opportunity with a low level of conviction, and I updated this rating on September 3, 2026.