AI Plays besides NVIDIA

[Nbadan]

This thread is not intended as investment advice but is merely for infotainment purposes only. You assume any and all risks should you choose to invest based on information in this thread. Neither the creator of this thread nor the owners of Spurstalk are responsible for bad investment decisions

That said....NVDIA isn't the only AI game in town and we aren't talking about the big 7....we will cover each of those in their own thread....so what are other AI plays were investors should be looking in 2024 and beyond?

[Nbadan]

Let's look at ....

Marvell Technology

Another semiconductor stock in this week's list is Marvell Technology (MRVL). The company managed to surpass analysts' expectations for the fiscal second quarter, even as revenue declined compared to the year-ago period. Management expects sequential revenue growth to accelerate in the fiscal third quarter, fueled by AI and cloud infrastructure.

In reaction to the results, Deutsche Bank analyst Ross Seymore reiterated a buy rating on MRVL stock with a price target of $70. The analyst noted that the company delivered a modest top-line beat and in-line outlook, with solid acceleration in AI-related applications offsetting macro-related weakness.

"Overall, we continue to believe MRVL has a compelling portfolio of infrastructure products that address powerful secular growth trends in AI/Cloud (electo-optics & significant custom compute), 5G and Automotive," said Seymore.

The analyst thinks that Marvell's infrastructure products, coupled with an eventual cyclical recovery in the storage, wired and on-premise businesses, would help in significantly accelerating the company's growth heading into calendar year 2024.

Seymore holds the 9th position among more than 8,500 analysts tracked on TipRanks. His ratings have been profitable 75% of the time, with each rating delivering an average return of 24.2%

[Nbadan]

Marvell's AI Chips Fail to Provide a Big Boost to Growth, but That Could Change Soon

A few months ago after the initial Nvidia AI pop, I said to be leery of chasing Marvell stock after it ran higher on AI optimism, too. The earnings update for the second quarter of fiscal 2024 (for the three months ended in July 2023) just proved why.

Marvell reported revenue of $1.34 billion in its second quarter, a 12% year-over-year decline and just a 1.5% quarter-over-quarter increase. Revenue is expected to be a bit higher in the third quarter, $1.4 billion at the midpoint of guidance, as Marvell keeps working through enterprise customer weakness (non-cloud data centers) due to excess inventory of some components and economic concerns (customers are in cash conservation mode this year).

[Nbadan]

Profitability wasn't so hot either. Net loss under generally accepted accounting principles (GAAP) was $208 million in Q2, or $290 million in net income on an adjusted basis. The difference between the two profit metrics is primarily non-cash depreciation and amortization expenses of $75.5 million and $272 million, respectively (from a string of acquisitions made a few years ago), and stock-based compensation of $153 million. Free cash flow (FCF) in the quarter was $1 million, compared to positive FCF of $105 million in the first quarter.

[Nbadan]

Marvell's chips used in the latest generative AI servers are off to a hot start. Come calendar year 2024, Marvell should be sitting on a new product line worth well in excess of $1 billion for the full-year period. That's significant, given that Marvell's all-time peak annual revenue was just shy of $6 billion in 2022. The key to the company surpassing that previous peak will be the rest of the data center and enterprise computing business solidifying going forward. Ongoing 5G network construction would help, too.

With Marvell still anticipating just a moderate uptick in sales activity in Q3, it's still hard for me to say it's time to buy. Shares trade for an elevated 60 times trailing-12-month free cash flow, although part of that high price tag has to do with depressed profitability during the ongoing semiconductor downturn. Perhaps by the fourth quarter, or early next year, things will start to heat up again. For now, Marvell is maybe a dollar-cost-average candidate for investors thinking generative AI is the real deal long-term, but prudence and patience could pay off here ahead of a more robust rebound in the business.

[Nbadan]

Here is what a sample of analysts are saying about Marvel

Morgan Stanley cut the price target on Marvell Technology from $68 to $65.
Morgan Stanley analyst Joseph Moore maintained an Equal-Weight rating.
B. Riley Securities raised the price target on Marvell Technology from $75 to $77.
B. Riley Securities analyst Craig Ellis maintained a Buy rating.
Oppenheimer analyst Rick Schafer, meanwhile, reiterated Marvell Tech with an Outperform and maintained a $70 price target.
Needham analyst Quinn Bolton reiterated Marvell Tech with a Buy and maintained a $65 price target.
Rosenblatt analyst Hans Mosesmann reiterated Marvell Tech with a Buy and maintained a $100 price target.

[Nbadan]

The rise in ultra-fast, high-capacity GPU-based servers for AI will also require a massive investment in networking, as large numbers of high-capacity switches and optical communications processors will be needed to quickly transmit huge amounts of data, both within and between AI data centers.

On the networking side, those investments should benefit Marvell's PAM4 digital signal processing (DSP) interfaces for lighting-quick data transmission at speeds of 800 Gbps and above, along with its data center interconnect products and Teralynx 10 Ethernet switching platforms.

[Nbadan]

n addition to networking, Marvell also has a growing ASIC business, in which Marvell collaborates with third parties to design custom chips built for a specific purpose. Aside from the large generalized GPU accelerator companies, leading cloud providers are now designing custom ASICs for AI as an alternative, and Marvell is set to benefit from the growth of these types of custom AI ASICs as well.

In early June, JPMorgan Chase analyst Harlan Sur issued a bullish note on Marvell, identifying that it has about 15% of the custom ASIC market -- a market that Sur projects will grow 20% annually over the medium term.

That optimism appeared validated just one week later, when the Taiwan-based Liberty Times reported that Marvell had won a new ASIC contract with Amazon Web Services for Amazon's next-generation accelerators.

[Nbadan]

With the AI segment apparently primed for multiyear hypergrowth and the potential for a recovery in the other core businesses, Marvell's stock soared in 2023's first half.

Still, I'd remain cautious on buying Marvell today after its strong run. The company is still only operating around breakeven, with lackluster total-company growth. So if Marvell delivers anything other than strong growth in the quarters ahead, the stock could pull back. That being said, it's certainly an AI name to watch over the medium and long term.

i currently do not own any Marvell shares

[Nbadan]

Palo Alto Networks

Next up is cybersecurity provider Palo Alto Networks (PANW), which reported better-than-anticipated fiscal fourth-quarter earnings. Revenue grew 26% year-over-year to $1.95 billion but slightly lagged estimates.

BMO Capital Markets analyst Keith Bachman, who ranks 584th out of over 8,500 analysts on TipRanks, noted that the company's fiscal 2024 guidance of 19% to 20% year-over-year billings growth and an 37% to 38% adjusted free cash flow (FCF) margin was better than expectations of mid-teens billings growth and a FCF margin in the mid-30% range.

Bachman thinks that the trend of consolidating solutions with leading security vendors will continue as the threat landscape evolves and as generative AI emphasizes the need for data aggregation. He added that implementing a consolidated portfolio enhances the prospects for real-time threat detection and remediation.

The analyst highlighted that customers are increasingly adopting each of PANW's three platforms (Strata, Prisma and Cortex), as they look for integrated solutions and unified data models. He increased his price target to $275 from $235 and reiterated a buy rating on Palo Alto.

"We believe that the strength of PANW's portfolio and the consolidation of spend are key drivers of PANW's long-term targets and net new NGS ARR [next-generation security annual recurring revenue] growth," said Bachman.

[Nbadan]

Palo Alto Networks had a vision back in 2018, spearheaded by then-new CEO Arora, to get the company back on track for a coming explosion in cybersecurity demand. Specifically, the need for new types of network protection, including public cloud (use of a remote data center accessed via the internet) and automation of security tasks using AI, were what Palo Alto Networks was hitching its future to then.

Suffice it to say, it's paid off. The company now touts itself as a full-fledged platform for all types of cybersecurity. It was a bumpy road, one that included over a dozen acquisitions of small start-ups along the way, culminating with the purchase of Cider Security, announced in late 2022. But since 2018, revenue has gone from about $2.5 billion to $6.9 billion in fiscal 2023, free cash flow remained positive all along the way, and net income is fast on the mend after the transformation.

[Nbadan]

As for the final numbers for fiscal 2023, revenue increased 25% year over year. Full-year GAAP net income was $440 million, compared to a loss of $267 million in 2022. And free cash flow (FCF) was $2.6 billion (a very good FCF profit margin of 38%), up a whopping 47% from 2022.

[Nbadan]

Going forward, Palo Alto Networks thinks its cybersecurity platform will continue to be a winning formula. Indeed, the amount of digital data out there has exploded, as has the number of ways for employees to access it. Security priorities will revolve around expanding zero-trust architecture (rooted in the "never trust, always verify" principle that requires users to constantly validate their credentials), security built directly into apps themselves during the coding process (the rationale behind that latest acquisition of Cider Security), and more automation of real-time monitoring.

The last five years have been good, and it looks like management thinks they'll continue. Its goals through fiscal 2026 (which ends in July 2026) are for revenue to increase an average of 17% to 19% each year. And along the way, the adjusted operating profit margin is expected to expand to a range of 28% to 29%, up from 24% this past year.

[Nbadan]

Of course, Palo Alto Networks will need to execute this plan to justify making a buy today. Its shares now trade for 30 times trailing-12-month free cash flow, a bit on the high side compared to where it has traded the last five years. And as good as the outlook is going forward, Arora and the top team do seem to be expecting a modest cooldown from recent growth.

[Nbadan]

Palo Alto Networks' Debt And Its 36% ROE

We think Palo Alto Networks uses a significant amount of debt to maximize its returns, as it has a significantly higher debt to equity ratio of 3.20. While its ROE is no doubt quite impressive, it could give a false impression about the company's returns given that its huge debt could be boosting those returns.

[Nbadan]

There's no denying that Palo Alto Networks is an expensive stock. Shares are currently priced at 13 times annual sales, which is higher than investors have seen since well before the pandemic. Microsoft is trading for 12 times sales, for context, even though the company is much more profitable and owns a dominant market position across several growth niches in the software-as-a-service space.

You could still see excellent returns from owning Palo Alto over the next several years, if the company can extend its current positive sales and profit margin momentum. Most Wall Street pros are expecting this to happen in 2023 and beyond. On average, they predict sales will rise 25% as earnings improve to $4.27 per share from $2.52 last year.

[Nbadan]

e cybersecurity market is estimated to grow at a CAGR of 13.8% between 2023-2030. We think PANW will be among the primary benefactors of the cybersecurity opportunity; this quarter, the company reported over $1B in bookings for FY23 in SASE and Cortex, and over $500B in Prisma Cloud ARR in Q4. We're specifically constructive on PANW growing its larger deal bookings, with the number of deals over $20M increasing 43% Y/Y, outpacing smaller deals under $10M, which grew at 37% Y/Y. PANW growing large deal sizes highlights the company's relative resilience to macro headwinds.

I currently own 4 shares of PANW

[Nbadan]

Palantir (PLTR)

Palantir operates AI data mining platforms for government agencies and enterprise businesses. Gotham, Palantir’s government platform, finds patterns in disparate data so intelligence teams can locate and respond to terrorism threats. Unconfirmed sources say Gotham played a role in capturing Osama bin Laden in 2011.

Businesses use Palantir Foundry to house, transform and manipulate organizational data to streamline processes and make better decisions.

Palantir went public in 2020 and recently reported its first profitable quarter (fourth quarter 2022).

[Nbadan]

Is Palantir Stock an Undervalued Gem?

While Snowflake and MongoDB trade at meaningful premiums to rest of the field, Palantir trades in line with Datadog, and ServiceNow is not too far behind. All of these companies are investing heavily in AI, and it's interesting to see which names are garnering more support from the markets.

To me, upon looking at the analysis above, the more interesting takeaway is that Palantir essentially sits right in the middle of this cohort. But as a longtime Palantir bull, I know there is far more to the story. In the next section, I'll dig into whether Wall Street sees potential for Palantir.

[Nbadan]

To summarize, Ives estimates that Palantir will reach $5 billion in revenue by fiscal 2027, and he uses a 10x multiple to derive a $25 price target. Although the forecast of a more than 60% upside from Palantir's current trading levels can be tempting, Rosenbaum makes a great point that the report lacks details supporting these future profitability estimates. In other words, while $5 billion in sales by 2027 is impressive, what is the opportunity cost?

More recently, Morgan Stanley downgraded Palantir stock to $8, which implies a 40% drop from current trading levels. According to several public reports, the bank's downgrade anchors on the gap between Palantir's AI progress and how much it is moving the needle for the company.

This is an interesting stance because Palantir's management has highlighted how much demand the company is experiencing from its AI products. For example, per the company's most recent shareholder letter, investors learned that the company's latest AI development has been deployed across 100 organizations, and the company's current pipeline has 300 or more identified opportunities. But Morgan Stanley could have a point here. Despite the impressive stats above, Palantir is only guiding toward approximately 16% top-line growth this year, far less than other notable beneficiaries of the AI boom such as Nvidia.

Although it would be nice to see a more pronounced reflection of AI demand in Palantir's financials, my view is that Morgan Stanley's stance is a bit shortsighted. Some companies, like Nvidia, will witness a higher surge in demand sooner than others due to how heavily its customers rely on its chip products. However, a company like Palantir is in a more unique position.

While the company sells its software across a variety of industries and end-markets, investors should understand a couple of nuances. First, the sales cycle for enterprise software can be quite long. It could easily take weeks or even months of demoing the products before a formal deal is put in place. Moreover, while the notion of digital transformation and where AI sits in that equation is increasingly becoming a top priority for executives, that does not necessarily mean that all potential customers currently have an identified use case in which AI is ready to be deployed.

[Nbadan]

If you invested in Palantir stock earlier this year when it was trading around its lows, it's probably not a bad idea to take some gains and recoup your initial investment, given how much the stock has run up. However, as a long-term investor, my suggestion would be to hang on to the bulk of your position. Taking short-term profits at the expense of long-term gains is not a strategy that I suggest.

As of now, the initial demand around Palantir's AI suite is encouraging, but I think investors have yet to see the full potential and impact this can have on its financial profile. For this reason, I do not think the AI hype is fully reflected in Palantir stock, despite its somewhat rich valuation when compared to other compe ors.

While Morgan Stanley's research suggests a more conservative price target, I don't believe that waiting around for a massive pullback is prudent. My suggestion would be dollar-cost averaging into Palantir over time and remaining vigilant around the company's growth prospects.

[Nbadan]

Despite massive government contracts, Palantir has shown fundamental issues with its margins and profitability, consistently spending more than it makes. That’s a recipe for long-term value destruction, which is unfortunately what this meme stock has shown.

Perhaps things are changing, and that could certainly be the case. Palantir has continued to show vigorous growth, pumping out top-line revenue growth of 47% in 2020 and 41% in 2021.

[Nbadan]

The stock surged from $4.30 to $20 based on hype around the company’s potential to monetize artificial intelligence. While an obvious catalyst for growth reaccelerating, I’m skeptical as to Palantir’s ability to capitalize on this specific secular growth trend.

A positive call from Wedbush’s Dan Ives boosted the stock 15% in days. While bullish sentiment prevails, the company’s valuation of roughly 16-times sales ought to raise caution, especially for those who take fundamentals seriously.

Parsing out how much of Palantir’s valuation is still hype, versus how much is fundamental long-term cash flow growth, is difficult to do. However, it’s my view that the company’s valuation reflects a considerable amount of hype right now, and in the absence of revenue growth acceleration, this is a stock with significant downside risk from here.

[Nbadan]

Peter Thiel’s Palantir is seeing ‘unprecedented’ demand for its military A.I. that its CEO calls ‘a weapon that will allow you to win

AIP stands for Artificial Intelligence Platform. According to the company’s site, the tool can be used by militaries to tap the kinds of AI models that power ChatGPT to aid in battlefield intelligence and decision-making. A demo video shows how the platform can display and analyze intel on enemy targets, identify potentially hostile situations, propose battle plans and send those plans to commanding officers for execution.

The video emphasizes that the platform is safe and secure, allowing a client to decide what data the models can see and what they “can and can’t do on behalf of humans.”

“If you wheel these technologies correctly, safely, and securely,” Palantir Chief Executive Officer Alex Karp said during Monday’s call, “you have a weapon that will allow you to win, that will scare your compe ors and adversaries.”

Karp described the boom in large language models as a revolution “that will raise ships and sink ships.” Demand for AIP is like “nothing I’ve ever seen in 20 years of being involved in Palantir,” he said. “We are reorganizing our efforts aggressively to capitalize.”

Karp also said that while Palantir has had conversations about the platform with “hundreds” of potential partners, details like pricing and terms are still being decided. Palantir recently refocused its engineering teams and other resources around AI to meet demand, and are “running hard” at the opportunity, Karp told analysts.Play Video

I currently do not own any palantir

[Nbadan]

Adobe (ADBE)

Adobe makes software for content creation, marketing, data analytics, do ent management, and publishing. Its flagship product, Creative Cloud, is a suite of design software sold via subscription.

In 2022, Adobe announced new AI and machine learning (ML) capabilities in its Experience Cloud product, a marketing and analytics suite. These advancements include predictive capabilities that help sales and marketing teams understand how the different facets of marketing campaigns affect customers' buying decisions. They can use that information to optimize campaigns and their budgets.