Taiwan Semiconductor Manufacturing (TSM -6.68%), popularly known as TSMC, is having a forgettable 2025 so far despite starting the year on a bright note. Shares of the foundry giant have slipped by more than one-third from the 52-week high they achieved on Jan. 24.

TSMC’s pullback is a result of the overall negativity in tech stocks on the back of the tariffs imposed by the Trump administration. It is feared that the tariff war will lead to an increase in manufacturing costs for technology companies that make products outside the U.S., bumping up the cost of deploying artificial intelligence (AI) data centers and forcing tech giants to rein in their spending.

Additionally, tariffs are expected to negatively affect the U.S. economy’s growth, which explains why there has been an increase in the probability of a recession. All these factors have weighed on TSMC stock this year.

However, a closer look at the company’s sales in the first two months of the year suggests that the stock could come out of the rut it is in. Specifically, it won’t be surprising to see TSMC stock stepping on the gas once again following the release of its 2025 first-quarter earnings report on April 17.

Let’s see why TSMC is poised to deliver stronger-than-expected results and guidance this month.

TSMC’s sales are growing at a nice clip

Taiwan Semiconductor’s revenue in the first two months of 2025 increased at an impressive pace of 39% when compared to the first couple of months of last year. At this pace, TSMC seems well on its way to exceeding its revenue guidance for Q1 2025.

When TSMC released its fourth-quarter 2024 results in January this year, the company guided for $25.4 billion in revenue for Q1 at the midpoint of its range. That would translate into a year-over-year increase of 34%, a big improvement over the 13% revenue growth it delivered in the year-ago period. However, Taiwan Semi’s growth trajectory for the first two months of the year indicates that it could end up outperforming its own expectations.

Meanwhile, earnings should also grow at a terrific pace, considering that TSMC is expecting a year-over-year jump of 5.5 percentage points in its operating margin. Not surprisingly, analysts are expecting a 49% increase in Q1 earnings from the prior-year period to $2.05 per share, though it could do better than that considering the robust AI chip demand.

TSMC’s surging sales can be attributed to the rapidly growing demand for AI chips that are now being deployed in multiple applications ranging from data centers to smartphones to personal computers (PCs) to automotive. Nvidia, which is one of TSMC’s top customers, reported recently that it is witnessing an unprecedented demand for its latest generation of Blackwell AI graphics processing units.

Taiwan Semiconductor fabricates the Blackwell GPUs designed by Nvidia. It has been focused on aggressively increasing its AI chip production capacity to fill Nvidia’s orders. Reports suggest that Nvidia has cornered more than 70% of TSMC’s advanced chip packaging capacity to meet Blackwell demand. What’s more, TSMC’s advanced chip packaging module shipments are reportedly increasing by 20% every quarter, which is why the company is looking to add two more facilities to boost supply.

Nvidia is expecting its revenue to jump 65% in the current quarter, and it gets most of its revenue from selling AI data center chips. Throw in the supply chain improvements that TSMC is making, and it’s easy to see why there’s a good chance that its quarterly performance and guidance could crush Wall Street’s expectations.

Meanwhile, other AI chip companies such as Broadcom and Marvell Technology have also called for outstanding growth in their sales. Broadcom and Marvell are benefiting big time from the rapidly growing demand for custom AI processors, which they design and TSMC manufactures. Similarly, another TSMC customer — Advanced Micro Devices — is witnessing an uptick in demand for central processing units that power personal computers (PCs), a market that’s getting a boost thanks to generative AI.

The future seems bright for Taiwan Semiconductor, as it is in a solid position to make the most of the secular growth of the chip market thanks to AI.

The stock is too attractive to ignore right now

TSMC stock’s recent pullback means that it can now be bought at under 25 times trailing earnings, while its forward earnings multiple of less than 19 points toward robust growth in the bottom line. These multiples are cheaper than the Nasdaq-100 index’s price-to-earnings ratio of around 29 (using the index as a proxy for tech stocks).

Analysts are expecting a 29% increase in TSMC’s earnings in 2025. Even better, analysts have been increasing their earnings growth expectations for the next couple of years.

TSM EPS Estimates for Current Fiscal Year data by YCharts.

However, there is a good chance that TSMC’s growth could be better than analysts are expecting. The company is forecasting its revenue to increase at a compound annual growth rate (CAGR) of 20% for the next five years. That’s why investors looking to add an AI stock that could deliver healthy long-term gains and that’s trading at an attractive valuation right now can consider loading up on TSMC before it starts soaring.

With the Nasdaq Composite (^IXIC 2.61%) down roughly 12% from its highs, many tech-related stocks have sold off heavily.

Making matters worse, some of these stocks just reported earnings that were less than perfect (according to the market), giving them a double whammy of negativity. Add in the uncertainty around tariffs at the moment, and the market has sold off numerous otherwise high-quality businesses out of fear.

One company squarely at the intersection of this trio of headwinds is global e-commerce enabler Global-e Online (GLBE 0.47%)

However, while the market has sent Global-e shares down 42% from their 2025 highs, I believe now is the time to buy the spectacular tech stock. Here are four reasons why it is a promising investment today.

1. Global-e Online is the leading force in a massive market

Selling products globally is often such a complicated task that most small businesses (and even some enterprise-sized companies) either can’t do it effectively, or don’t believe it is worth the hassle to even try.

That’s where Global-e Online’s end-to-end, global e-commerce platform takes over. Helping merchants in 30 countries (and counting) sell to over 200 countries across the globe, the company offers an array of solutions, including:

• Local pricing in 100-plus currencies
• Over 150 payment options
• Shipping options from more than 20 providers, along with local returns
• Messaging in over 30 languages
• Guaranteed calculations for local import duties and tariffs
• Zero-risk payment fraud management and assistance
• Know-how and data on each local market

Just how complex are these solutions?

Despite being one of the leaders in the e-commerce realm, Shopify chose to invest in and partner with Global-e rather than build out its own cross-border solutions. Powered by Global-e’s platform, the two combined to create Shopify Management Markets, which lets interested merchants in Shopify’s ecosystem sell in foreign markets. With 10,000 merchants using the service in just 18 months after its launch, it seems there is plenty of interest in doing so.

While businesses can try to go it alone when they sell internationally, they may be leaving money on the table. Merchants that switched to Global-e’s platform averaged a 40% uplift in international traffic conversion.

Global-e’s merchants are growing their gross merchandise volume by four to five times faster than the global e-commerce growth rate of 8% in 2024, demonstrating the company’s growth potential.

Increasing revenue by 32% in 2024 and guiding for 25% growth in 2025, Global-e should continue to rapidly gain share of a target addressable market that it believes is worth $3 trillion.

2. Tariffs and Global-e

While this guidance for 25% sales growth in 2025 was enough to keep me happy, it didn’t impress the market. Worried about the impact higher (or new) tariffs in the United States could have on Global-e’s business, the market seemed to take a sell-first, ask-questions-later stance on the stock.

However, co-founder and president Nir Debbi downplayed the impact of these tariffs during the company’s fourth-quarter earnings call — even hinting that they could be a long-term benefit, stating:

We do believe that there will be uncertainty. It might affect short-term consumption. But overall, in the longer term, we do expect it will behave the same way we’ve seen in Brexit, where overall, it created much more demand for our services.

Put together, there will probably be short-term troubles, but they may create long-term value. To me, that’s a Foolish opportunity.

3. Improving margins hint at a wide moat

Lost amid Global-e’s “disappointing” Q4 results was the fact that the company reached break-even profitability for the first time in the quarter and expects to remain profitable going forward. Over the last two years, the company’s margins have rapidly improved, serving as a testament to the wide moat it is building around its operations.

GLBE Gross Profit Margin and Profit Margin (Quarterly) data by YCharts.

Global-e’s net dollar retention (NDR) rate of 123% or higher over the last four years also reinforces the idea that the company is building a wide moat. Measuring how much existing customers increase their sales from one year to the next, an NDR consistently above 120% shows that Global-e’s solutions are almost a “no-brainer,” and merchants are happy to spend more each year.

4. An all-time low valuation

Best yet for investors, despite Global-e’s leadership position, improving margins, and widening moat, the company trades at an all-time low valuation.

GLBE P/S and P/FCF Ratio data by YCharts.

Trading at 37 times free cash flow (FCF) — just slightly north of the S&P 500‘s (^GSPC 2.13%) average price-to-FCF ratio of 32 — Global-e’s growth (which is multiples higher than the market’s) appears cheaply priced.

Already one of my core holdings, I will be looking to add to Global-e Online soon as the company continues to forge its path as the leading global e-commerce enabler.

Stock splits have an interesting narrative. During bull markets, investors can’t get enough of stock-split stocks as prices rise in anticipation for upcoming splits. This conflicts with the actual reality on the ground.

In actuality, stock splits have no impact on a company’s financial performance. After a forward stock split (where a company’s shares outstanding grow) or a reverse stock split (where shares outstanding contract), a company’s total shares outstanding will change, but it is still the same old business as the day before.

This doesn’t stop investors from getting enamored with stock-split stocks during a bull market, though.

Now, with the market in a correction, investors have fallen out of love with stock-split stocks. Does that make them a potential buying opportunity for your portfolio? Here is one stock-split stock to buy and one to avoid in the month of March.

Chipotle: Trading at its cheapest valuation in years

Chipotle (CMG 2.67%) split its stock 50-to-1 in 2024, bringing its share price down to a more manageable buying price for individual investors. As of this writing, its stock trades at a price of $50 and is in a 27% drawdown. Investors are concerned about a slowdown in restaurant spending at the moment, which is causing the sector to struggle to start 2025.

The underlying business looks just fine. In 2024, Chipotle’s revenue grew 14.6% to $11.6 billion. This was driven by opening new restaurants and increasing same-store sales of 7.3% compared to 2023. Operating margin was 16.9% in the year, up from 15.8% in 2023.

Add everything together, and Chipotle’s earnings per share (EPS) grew 24.7% year-over-year in 2024.

With plenty of room to grow its store count in North America — and eventually globally — I believe Chipotle can keep up this strong EPS growth for years to come.

After the stock’s drawdown, Chipotle trades at a price-to-earnings ratio (P/E) of 45. While this doesn’t look dirt cheap, it should come down quickly with how fast Chipotle is growing its EPS and is at its lowest level in the last five years, excluding the March 2020 stock market panic.

Investors who buy Chipotle stock today should do just fine holding over the long haul.

Sirius XM’s losing business model

SIRI data by YCharts

Forward stock splits are a sign a company’s business is succeeding. The price of your stock rises generally when the business does well. Reverse stock splits mean the opposite, which is why Sirius XM Holdings (SIRI 3.85%) recently implemented a reverse split. A leader in satellite radio, Sirius XM is struggling to pivot its business to modern digital audio streaming.

In 2024, revenue fell 4% to $6.6 billion. Free cash flow of around $1 billion is at its lowest level in 10 years. Its acquisition of Pandora Music has gone nowhere, and it is investing in expensive podcast licensing deals.

Despite prior years of success, Sirius XM is getting lapped by modern competitors like Spotify and YouTube, which are gaining millions of new users every year. Sirius XM’s ranks of users are moving in the wrong direction.

Dangerously, Sirius XM carries more than $10 billion of long-term debt on its balance sheet. If its free cash flow keeps moving in the wrong direction, this company could be headed for more trouble in the next few years, even with the stock already down 71% from all-time highs. Avoid buying Sirius XM stock for your portfolio in March.

Bill Ackman is the founder and CEO of Pershing Square, a hedge fund with around $12.9 billion in assets under management. That includes billions of dollars invested in popular stocks like Alphabet, Nike, and Chipotle Mexican Grill.

On Friday, Feb. 7, Ackman took to social media platform X (formerly Twitter) to reveal that Pershing Square has accumulated 30.3 million shares in Uber Technologies (UBER 3.12%), which operates the world’s largest ride-hailing platform. Based on Uber’s stock price of $74.60 at the close on Friday, it’s now Pershing’s largest position with a value of $2.2 billion.

Ackman believes Uber is trading at a discount to its intrinsic value, but I think the stock could deliver significant upside over the long term for another reason — autonomous vehicles, which have the potential to transform the company’s economics.

Image source: Getty Images.

Uber values the autonomous opportunity at $1 trillion

Uber serves over 171 million people every month across its ride-hailing, food delivery, and commercial freight services. The company accepted a record $162.7 billion in gross bookings during 2024 across all three segments — a figure that represents the full amount customers paid for every ride, food order, and commercial delivery.

Uber’s drivers earned a record $72.5 billion in 2024, which was the single largest piece of the $162.7 billion in gross bookings. After deducting driver costs and the money paid to restaurants for food orders, Uber was left with just $43.9 billion in revenue.

Therefore, if Uber could eliminate the enormous cost of its drivers, its revenue would organically soar because it would keep more of its gross bookings. That’s why the company is actively pursuing autonomous driving solutions, not only for its ride-hailing service but also for its food delivery and freight businesses.

Of course, some of the driver costs would be replaced by fees payable to the autonomous vehicle companies operating within its network, so Uber can’t eliminate that expense entirely. However, self-driving cars can operate 24 hours per day, seven days per week, with few ongoing expenses, so they will still be significantly better from a financial perspective.

Uber could also buy a fleet of autonomous vehicles and operate them itself, which means it could pocket the entire gross booking from every ride-hailing trip. Tesla is expected to sell its Cybercab robotaxi for around $30,000, which might be an attractive option. However, it would involve a big change to Uber’s business model, because it has always relied on drivers to supply their own cars (and as things currently stand, it will rely on its partners to supply autonomous vehicles).

Nevertheless, Uber CEO Dara Khosrowshahi thinks autonomy presents a staggering $1 trillion opportunity for the company in the U.S. alone.

Partnerships with autonomous titans like Waymo and Nvidia

Khosrowshahi says Uber is spending an enormous amount of organizational energy executing on its autonomous strategy, even though it might be a few years before self-driving cars are widely available. The company has inked partnerships with over a dozen companies developing autonomous solutions, including Alphabet’s Waymo, WeRide, Motional, Serve Robotics, and even Nvidia.

Waymo is ahead of the pack right now because it’s already completing over 150,000 paid autonomous trips every week across Phoenix, San Francisco, and Los Angeles (many of which are through Uber). Later this month, Uber customers in Austin will be able to hail an autonomous Waymo, and the partnership will expand into Atlanta later in 2025.

Uber is also working with Serve Robotics to launch over 2,000 autonomous food delivery robots this year, which will operate exclusively on Uber Eats in parts of California and Texas. During the fourth quarter of 2024, Uber started offering autonomous food delivery in Austin and Dallas through a different partner called Avride.

Then, there is the interesting deal that Uber signed with Nvidia in January. Since Uber facilitated over 12 billion trips last year, it has mountains of useful data that could be used to train the artificial intelligence models underpinning most autonomous software. The company wants to use Nvidia’s Cosmos platform and DGX Cloud to process that data and help its autonomous partners accelerate their journey to commercialization.

The faster autonomous vehicles are widely deployed, the faster Uber can unlock the significant cost savings I talked about earlier. Plus — and this is just speculation on my part — Uber could possibly unlock a new revenue stream by charging its partners for the use of its valuable data.

Ackman and Pershing Square could enjoy significant upside in Uber’s stock

After deducting operating expenses like marketing and research and development from Uber’s $43.9 billion in 2024 revenue, the company was left with a net income (profit) of $9.8 billion, which translated to $4.56 in earnings per share (EPS). That places Uber stock at a price-to-earnings (P/E) ratio of 16.3, which is a whopping 51% discount to the P/E ratio of the Nasdaq-100 technology index (33.6).

Uber stock is remarkably cheap from that perspective, so it’s easy to see why Ackman thinks it’s a bargain right now. However, the company benefited from one-off tax benefits worth $5.7 billion in 2024, so its actual earnings were much lower in reality. In fact, according to Wall Street’s consensus forecast (provided by Yahoo), Uber’s EPS could shrink by 46% this year to $2.44 as those benefits disappear.

That places Uber stock at a forward P/E ratio of 30.5, but that’s still quite attractive. If we assume the P/E ratio of the Nasdaq-100 remains constant, the stock would have to climb by 10% in 2025 just to trade in line with its big-tech peers. Plus, Wall Street is forecasting 36% EPS growth for Uber in 2026, which is likely to drive even further upside in the stock in the medium term.

However, for the reasons I highlighted earlier, autonomous driving might be the best reason to own Uber stock for the long term. If it truly does become a $1 trillion tailwind as Khosrowshahi expects, the stock is likely very undervalued right now.