Uh-Oh! Alphabet Is Selling Its Stake in 2 Historically High-Flying Artificial Intelligence (AI) Stocks

Five weeks ago, what can arguably be described as the most important data release of the third-quarter occurred — and I’m not talking about the July inflation report.

No later than 45 calendar days following the end to a quarter (Aug. 14 in the latest instance), institutional investors with at least $100 million in assets under management are required to file Form 13F with the Securities and Exchange Commission. A 13F provides a clear-cut rundown of which stocks Wall Street’s brightest, and often richest, investors bought and sold in the latest quarter. Even though these snapshots are up to 45 days old when filed, it offers invaluable insight into the stocks, industries, sectors, and trends have the attention of money managers.

Image source: Getty Images.

However, 13F filings are a necessity for more than just investment institutions. Some of Wall Street’s largest and most-successful companies hold sizable stakes in other businesses, including Nvidia, and are required to file a 13F on a quarterly basis.

One such company, which has nearly $2 billion invested across 42 stocks, is Alphabet (GOOGL 0.38%) (GOOG 0.39%), the parent of search engine Google, streaming platform YouTube, and autonomous ride-hailing service Waymo, among other ventures.

Alphabet is a search engine juggernaut… and much more!

Most investors are familiar with Alphabet because of its world-leading Google search engine.

In August, Google accounted for a mammoth 90.48% share of worldwide search, per GlobalStats. In fact, Google has held at least a 90% monthly share of global internet search dating back more than nine years. This makes it the undisputed go-to for businesses wanting to target users with their message(s) and affords parent Alphabet unbelievable ad-pricing power.

You’re probably also familiar with YouTube, the second most-visited social media site on the planet, with approximately 2.5 billion monthly active users. The launch of Shorts — short-form videos lasting 60 seconds or less — in the latter half of 2020 provided an additional opportunity for the company to capitalize on ad revenue.

I’d be remiss if I didn’t also mention that Alphabet’s Google Cloud is the world’s No. 3 cloud infrastructure service platform by spending, with 10% of the market, per Canalys, as of the June-ended quarter. Following years of losses, Google Cloud shifted to recurring profits in 2023 and hasn’t looked back. Since cloud-service margins often outpace advertising margins, this segment should be a key cash-flow driver for Alphabet throughout the remainder of the decade.

But Alphabet is also an investor. It has nearly a quarter of its portfolio invested in development, security, and operations software developer GitLab, and more than 16% of invested assets tied up in Arm Holdings. Arm generates its revenue via licenses and royalties from chipmakers that use its designs to produce central processing units, graphics processing units, and other hardware.

However, it’s not what Alphabet holds that’s raising eyebrows. Rather, it’s the two core artificial intelligence (AI) stocks it’s been selling in consecutive quarters.

Image source: Getty Images.

CrowdStrike Holdings

The first AI stock Alphabet’s investment managers have shown the door in back-to-back quarters is cybersecurity solutions provider CrowdStrike Holdings (CRWD 3.35%). After reducing its stake in CrowdStrike by a third during the March-ended quarter, Alphabet slashed its remaining position by another 50% (427,894 shares) in the June-ended quarter. CrowdStrike is now Alphabet’s fourth-largest position, down from No. 2, where it began 2024.

Valuation is the likely culprit behind this selling activity. Even though CrowdStrike has blown the door off of Wall Street’s consensus growth forecasts for years, it was trading at nosebleed price-to-sales (P/S) and forward price-to-earnings (P/E) ratios. Locking in gains was likely deemed a prudent move.

The more recent concern with CrowdStrike — albeit one that occurred in July, which wouldn’t be detailed by the latest round of Form 13Fs — is a botched update to its Falcon security platform that caused significant downtime for select industries and customers. It’s not uncommon for these snafus to cost cybersecurity companies revenue in the short run.

On the bright side, CrowdStrike’s error was self-inflicted and had nothing to do with a cyberattack. Prior to this July outage, businesses had demonstrated a willingness to pay a premium for CrowdStrike’s services given its history of protecting against breaches.

What’s more, CrowdStrike has mastered the art of the add-on sale within the cybersecurity arena. In less than seven years, it went from a single-digit percentage of its customers purchasing four or more cloud-module subscriptions to 65% of its clients using five or more cloud modules, as of the July ended quarter.

The cherry on top for CrowdStrike is that cybersecurity solutions have evolved into a basic need service. No matter how well or poorly the U.S. economy is performing, businesses with an online or cloud-based presence need to protect their data.

With the cloud-native Falcon platform only growing more effective over time, thanks to AI, Alphabet’s brightest investment minds may regret their decision to sell a collective two-thirds of their stake in CrowdStrike.

DexCom

The other AI stock that Alphabet’s investment team has been dumping for two straight quarters is medical-device giant DexCom (DXCM 0.83%). After shedding slightly north of 51% of its stake in the maker of continuous glucose monitoring systems (CGMs) in the first quarter, Alphabet sold another 42.1% of its position (753,836 shares) in the second quarter. This former No. 1 holding of Alphabet has slipped to No. 6 and now accounts for about 6% of invested assets.

Similar to CrowdStrike, valuation has long been the glaring red flag with DexCom. Although it’s been growing sales at a relatively steady rate of around 20% for years, DexCom’s P/S and forward P/E ratios have always been eyebrow-raising.

The more-pressing concern of late for DexCom is what might happen to its business as a result of glucagon-like peptide-1 (GLP-1) agonist drugs hitting the market. GLP-1 therapies have been shown to reduce weight in patients taking them, and obesity is a common co-morbidity for people with diabetes. Although studies have shown that patients on GLP-1 drugs are more likely to use CGMs, DexCom’s second-quarter operating results, which were reported in July (i.e., after 13Fs were filed), did little to quell these concerns.

DexCom lost 40% of its value in the blink of an eye after lowering the midpoint of its full-year sales guidance by about $250 million and highlighting a laundry list of challenges during the quarter. CEO Kevin Sayer pointed to lower revenue per user and the company restructuring its sales team as key reasons for this shortfall. In particular, Sayer believes a lack of sales team coverage in select geographic areas led to the company’s new patient count coming up short.

If there’s a silver lining here, it’s that DexCom is one of the two leading providers of CGM devices, and should, in theory, benefit from the sheer number of people diagnosed with diabetes in the U.S. and worldwide climbing.

DexCom can also lean on its various AI solutions to differentiate itself and provide value to its users. This includes everything from glucose control for its infusion pumps to helping users with dietary assessments to optimize their blood glucose levels.

But the nature of DexCom’s earnings miss is concerning, to say the least. Color me not surprised if Alphabet continues to purge DexCom from its investment portfolio in the September-ended quarter.