BUSINESS

Looking for help with today’s New York Times Wordle? Here are some expert hints, clues and commentary to help you solve today’s Wordle and sharpen your guessing game.

TheStreet aims to feature only the best products and services. If you buy something via one of our links, we may earn a commission.

Why we love this deal

With fall approaching, it’s a good time to spend more afternoons outside before the weather turns too cold. Pickleball can be an easy activity to organize with friends and family, especially when people may just want to play a few games for the social aspect instead of having to go for a more demanding sport or just sit and chat. Pickleball is a great way to get some movement in while still having fun, and having enough equipment for those spur-of-the-moment games with friends makes it easier to pull a group together, whether you’re heading to a local court or packing up for a weekend vacation.

The Aoborty Pickleball Paddle 4-Pack with Accessories provides enough for four people, plus a travel case to take it out to the court or toss it in the car for the weekend. It includes everything needed for a fun day at the court for just $40. Shoppers save 20%!

Aoborty Pickleball Paddle 4-Pack with Accessories, $40 (was $50) at Amazon

Courtesy of Amazon

Shop at Amazon

Why do shoppers love it?

This set includes six balls and four paddles with sweat-resistant grips, plus a carrying case for all of it. The balls include three indoor balls and three outdoor balls, so you can play in a variety of spaces depending on what’s accessible. Each paddle weighs only 7.7 ounces, making them light enough to play for long periods of time and also making them easier to handle, especially for younger players. They have a fiberglass face and a graphite shaft designed to balance power and control. The 4.5-inch grip has an ergonomic, cushioned handle with a preferred surface that wicks away sweat while you play, preventing the paddles from slipping out of your hand. 

Related: Skip gym memberships and reach your step goal with an affordable walking pad

The set is suitable for teens, adults, and older players, as well as beginners and experts alike, making this set useful for households of all types. Since it has four paddles, it can also include singles and doubles games, giving you the chance to just play one-on-one or swap out in doubles if you have bigger groups. These paddles are also USA Pickleball Association approved, allowing you to use them in competition, while also being suitable for less formal games at gyms and sports clubs. 

Details to know

Items: This set includes four pickleball paddles, three indoor balls, three outdoor balls, and a carrying case.

Lightweight: Each paddle weighs under 8 ounces, and includes a sweat-wicking comfort handle. 

USAPA approved: These paddles are USA Pickleball Association approved, allowing you to use them in competition. 

One reviewer wrote, “These paddles are a good quality for the price. The handle wraps are comfortable and seem like they will hold up well enough. They fit perfectly in my hand, and provided the perfect amount of grip. The balls are included, as well as a bag to carry it all. I could not be happier with the price and quality.”Another shopper wrote, “Hours of fun have been had already, and they are holding up great.”

Shop more deals

Portable Pickleball Net, $63 (was $70) at Amazon

Portable Pickleball Net and Accessories, $54 (was $60) at Amazon

Smartway USAPA Pickleball Paddle Two Set, $30 (was $37) at Amazon

Whether you’re looking for a fun activity to get some movement in or you want to find your new favorite sport, the Aoborty Pickleball Paddle 4-Pack with Accessories offers enough to get you and your friends started. The carrying case makes it easy to keep everything together, and it’s great to have the option to play singles or doubles on the court. At just $40, you get all these items for less than $4 per piece. 

Tom Cruise’s “Days of Thunder” co-star is angry about the casting of Anne Hathaway in the long-awaited sequel to the 1990 racecar movie hit.

Most companies would kill for the biggest problem Nvidia (NVDA) has.

The AI-chip giant just generated $96.2 billion in quarterly revenue, more than double its sales from a year earlier. Its Data Center division reached $89 billion, and Nvidia forecasts another $108 billion of companywide revenue in the current quarter.

Then management said something that alters the way investors should look at those numbers.

Nvidia’s earnings call predicted revenue would rise about 70% in the fiscal year ending January 2028. But the company also expects supply constraints to continue at least through the end of that fiscal year.

That’s an odd situation for a company that is already running at Nvidia’s scale.

Demand isn’t currently the obvious ceiling.

Supply is.

The difference matters because analysts are starting to think about numbers that would have seemed almost absurd just a few years ago. Raymond James analyst Simon Leopold thinks Nvidia could potentially generate $1 trillion in annual revenue in fiscal 2029.

That is well above current Wall Street expectations of just under $750 billion.

But Nvidia’s latest results suggest the path to such numbers could depend on something much more tangible than AI enthusiasm: whether Nvidia and its suppliers can churn out enough GPUs, memory, networking equipment, and complete computing systems to satisfy customers who are already lining up to buy them.

Nvidia is selling nearly twice as much as it did a year ago

First, just the sheer number of hardware units Nvidia is already shipping.

Fiscal second-quarter revenue increased 106% year over year and 18% sequentially to $96.2 billion.

Data Center revenue soared 117% to $89 billion.

That means Data Center generated about 92.5 cents of every dollar Nvidia made in the quarter.

The profit numbers are just as striking.

GAAP operating income reached $63.7 billion, up 124% from a year ago. Net income climbed 126% to $59.7 billion, while GAAP earnings per diluted share increased 128% to $2.46.

Nvidia also kept a 75% GAAP gross margin.

Those numbers matter for the supply story because Nvidia is easily turning demand into profit as it scales.

The company reported net income of nearly $60 billion on revenue of $96 billion in three months.

Related: Bank of America doubles down on Nvidia stock

And the company isn’t expecting the growth to slow this quarter.

Nvidia expects third-quarter revenue of $108 billion, plus or minus 2%.

That means Nvidia expects to make more revenue in one quarter than it did in the entire fiscal year ending in January 2024, when annual revenue was $60.9 billion.

Nvidia’s growth forecast comes with an unusual warning

The more telling number came when Nvidia looked beyond the next quarter.

Management provided a preliminary outlook for fiscal 2028, anticipating revenue growth of around 70%.

That’s a tremendous projection for a business that already makes hundreds of billions of dollars a year.

But maybe even more important than that forecast is the sentence that comes with it:

“We expect supply to remain a bottleneck at least through the end of fiscal year 2028,” NVIDIA Chief Financial Officer Colette Kress said during the Q2 FY2027 earnings call.

That tells investors something about the balance of demand for Nvidia’s chips and the company’s production capacity.

The company is forecasting less than 70% growth and warns that customers could disappear.

It is forecasting 70% growth while saying its ability to supply those customers remains constrained.

More Nvidia:

Nvidia just made a move Wall Street wasn’t ready for

Nvidia just locked down deal that changes AI race

Nvidia stock is doing something it hasn’t done in years

The demand is also getting broader.

Revenue from a category that includes AI-native companies, enterprises, and sovereign customers grew 138%, while hyperscaler revenue more than doubled from a year ago, said Nvidia.

That matters because one of the big bearish arguments that has swirled around AI stocks is concentration: A handful of huge technology companies are pouring huge amounts of money into infrastructure.

Nvidia’s numbers suggest demand is spreading beyond that initial group.

Nvidia just told Wall Street what could cap its next leg higherBloomberg / Getty Images

Nvidia’s next AI system makes the supply problem even bigger

Meanwhile, Nvidia is undergoing yet another major product transition.

Its Vera Rubin architecture is entering production, and Nvidia expects Rubin to represent roughly 20% of Data Center revenue in the third quarter, with supply ramping further afterward.

The change is important because it’s not just about discrete graphics processors that Nvidia is shipping anymore.

Its AI business is increasingly about full systems with GPUs, CPUs, networking, high-bandwidth memory, and other components that need to work together at massive scale.

That makes for a lot more complicated supply chain. One shortage can impact delivery of a whole system. A particularly important example is memory.

On its latest earnings call, Nvidia acknowledged what it called “extreme pricing conditions” in memory.

AI accelerators require high-bandwidth memory, or HBM, capable of moving enormous quantities of data rapidly enough to keep increasingly powerful processors working efficiently.

The more GPUs that Nvidia ships, the more memory the industry will need.

The same principle extends to packaging, networking, racks and power infrastructure.

So Nvidia’s challenge is more than just making another chip.

It needs to develop a full AI-computing ecosystem fast enough to meet extraordinary customer demand.

The $1 trillion Nvidia scenario depends on solving this problem

Here’s where Raymond James’ $1 trillion scenario comes in handy.

This is not a target, but rather a measure of what could be on the other side of Nvidia’s supply constraints.

MarketWatch has a current year revenue estimate for Nvidia of about $403.5 billion.

If revenue then grew at Nvidia’s prior 70% fiscal 2028 growth rate, annual sales would approach $686 billion.

Nvidia would need another 46% rise from around $686 billion to reach $1 trillion.

That would normally sound extreme. But Nvidia just grew quarterly revenue 106%. Its Data Center business grew 117%. And another major group of AI customers grew 138%.

Raymond James’ scenario is still substantially more bullish than consensus. FactSet estimates put fiscal 2029 revenue at below $750 billion.

That’s about $250 billion a year wrong.

That gap shows how much hinges on Nvidia’s ability to increase supply.

If demand stays strong and Nvidia can build and ship many more systems, today’s consensus estimates could be conservative.

If manufacturing capacity, memory availability, packaging or other infrastructure can’t ramp up quickly enough, Nvidia could be leaving huge amounts of potential revenue on the table.

Nvidia’s customers are creating another constraint

There’s another physical limitation that Nvidia can’t fix on its own.

Its customers need places to put all this equipment.

AI data centers need huge amounts of electricity, land, cooling equipment and networking infrastructure.

This makes Nvidia’s growth equation a little bit broader:

Chip supply + memory + servers + data centers + electricity = potential revenue.

The larger the scale of AI spending, the more important each of those variables becomes.

Nvidia CEO Jensen Huang has long argued that the world is moving from traditional general-purpose computing to accelerated computing and AI infrastructure.

The company’s financial results show the extent of that transition. But now the industry has to physically build the infrastructure to carry it. That’s why Nvidia’s supply warning is more notable than the usual quarterly shortage. Quarterly sales of $96.2 billion can mean very large amounts of revenue from small constraints.

A bottleneck equal to just 5% of Nvidia’s current quarterly revenue would represent roughly $4.8 billion.

At a hypothetical $700 billion annual revenue run rate, the same 5% would represent $35 billion.

The larger Nvidia becomes, the more expensive every constraint becomes.

Nvidia investors may be watching the wrong risk

For much of the past year, investors have been focused on whether companies such as Microsoft, Amazon, Alphabet, and Meta can keep spending extraordinary amounts on AI infrastructure.

But the results from Nvidia don’t take that risk away.

But they change the immediate question.

Nvidia just reported a 106% jump in revenue, a 117% jump in Data Center, and almost $60 billion in net income for the quarter. Management guided for about 70% revenue growth for the next fiscal year while simultaneously warning that supply will remain constrained.

The market responded.

Nvidia shares jumped 8.7% after earnings, adding about $442 billion in market capitalization in a single session.

The stock reaction was more than just another earnings beat.

Investors received evidence that Nvidia believes the expansion of AI infrastructure can remain strong well beyond the next several quarters.

This news raises the importance of the company’s production capacity.

One day there will be a plateau in AI demand. No company can double forever, especially one with annual sales in the hundreds of billions of dollars.

But recent numbers from Nvidia suggest it isn’t there yet.

Instead, the company is facing a much stranger problem.

It has customers paying enough to generate $89 billion of Data Center revenue in a three-month period, it expects about 70% growth in the next fiscal year, and it doesn’t believe the supply chain will ever fully catch up.

That trillion-dollar revenue debate is just an insane headline.

The more relevant fact for investors is what Nvidia needs to do before that number could ever become possible.

It has to build enough AI infrastructure to satisfy demand that is already there.

Related: OpenAI’s agents breached Hugging Face. Nvidia wants it.

TheStreet aims to feature only the best products and services. If you buy something via one of our links, we may earn a commission.

Why we love this deal

With less than a month left of summer, you might think that the time to shop for patio furniture has passed. However, it’s actually one of the best times to get a great deal on a new set. With retailers clearing out inventory for a new season, outdoor furniture is deeply discounted. That means it’s the perfect time to upgrade your setup or add a new set to your space.

The Elposun 4-Piece Rattan Patio Set is on sale for only $99 at Walmart, which is 43% off its regular price of $175. Even with a $40 shipping fee, it’s a steal. With a massive clearance deal, it’s a wonderful set for a small patio or a spacious deck.

Elposun 4-Piece Rattan Patio Set, $99 (was $175) at Walmart

Courtesy of Walmart

Shop at Walmart

Why do shoppers love it?

The Elposun patio set comes with four pieces, including a loveseat, two armchairs, and a coffee table. It’s a step up from the standard three-piece set, which typically only comes with two chairs and a table, providing more places to sit and lounge in an outdoor space.

Each piece is made of steel and wrapped in handwoven rattan. Not only do they look stylish, but they’re also durable. The set has an all-weather design that can hold up against the seasons, rain or shine. The loveseat and chairs come with cushions, all of which have zippers on them. Since you can remove the covers, you can easily wash them when needed. The coffee table is the perfect final touch, bringing the entire set together, almost giving the appearance of an outdoor living room. With a tempered glass top, it’s a sleek accent that’s also easy to clean with a simple wipe-down.

You can get the patio set in four colors: beige, black, gray, and navy.

Related: Walmart has an indoor-outdoor plush oversized rocking chair for 49% off

Details to know

Includes: A loveseat, two armchairs, and a table.

Weight capacity: 450 pounds for the loveseat and 225 pounds for the armchair.

Colors: Beige, black, gray, and navy.

One Walmart shopper said that this patio set has one of the best prices. They said they love the size, and the table cleans well. To make sure it lasts, the shopper said they keep it out of the rain, even though it’s all-weather. It holds up well, but they claim the seats can retain rainwater, so they let them drain out before they use them. Better yet, you can use patio covers to protect them from the rain or bring the cushions inside or put them in an outdoor storage box when inclement weather is expected.

Shop more deals

Lofka 4-Piece Rocking Chair Patio Set, $100 (was $181) at Walmart

Udpatio 4-Piece Patio Set, $90 (was $100) at Walmart

Lofka 4-Piece Rattan Patio Set, $110 (was $200) at Walmart

If you’ve been searching for a new patio set, the Elposun 4-Piece Rattan Patio Set is a great choice. It provides comfort with multiple seating options, and it’s only $99.

Enterprise AI has entered a new era. Organizations are rapidly moving beyond assistants that answer questions to autonomous agents capable of reasoning, invoking tools, accessing enterprise applications, coordinating with other agents, and completing multi-step business workflows with minimal human intervention.This shift represents a fundamental change in how software operates. Traditional applications execute predefined logic written by developers. AI agents, however, dynamically determine how to achieve an objective. They decide which tools to use, which APIs to call, what information to retrieve, and how to sequence actions based on context. That flexibility unlocks enormous business value, but it also introduces a new class of security risks.Much of today’s AI security discussion focuses on prompt injection, model vulnerabilities, and data leakage. These are important concerns, but they represent only part of the challenge. Once an AI agent has successfully authenticated and begins acting autonomously, traditional security controls provide very little visibility into whether it continues to operate safely.This is where enterprises need to adopt a new security mindset: runtime trust.Authentication establishes identity, not trustEnterprise security has traditionally relied on three foundational questions: Who are you, what can you access, and what actions are you authorized to perform. Identity providers, multi-factor authentication (MFA), role-based access control, and zero trust architectures answer these questions effectively for human users and conventional applications, and NIST’s zero trust guidance remains a solid reference point for how those principles are meant to work (NIST SP 800-207).AI agents introduce a different problem. An AI agent may legitimately authenticate using an enterprise identity, receive valid API credentials, and be granted access to systems like Microsoft 365, ServiceNow, Salesforce, or GitHub. From an identity perspective, everything appears correct. The real challenge begins after authentication: During execution, the agent continuously reasons, interprets objectives, invokes tools, retrieves information, and adapts its behavior based on new context, and security teams must determine whether those actions remain aligned with the user’s intent and organizational policy. Authentication verifies who an AI agent is. Runtime trust continuously verifies what it is doing.Enterprise AI is becoming an autonomous workforceModern AI agents increasingly interact with large language models (LLMs), Model Context Protocol (MCP) servers, retrieval-augmented generation (RAG) systems, vector databases, enterprise APIs, SaaS platforms, and internal knowledge repositories, as well as other AI agents. This interconnected ecosystem enables sophisticated automation but dramatically expands the attack surface: A single compromised tool, poisoned knowledge source, overly permissive API, or manipulated prompt can influence downstream decisions across an entire workflow, and unlike traditional software, these risks evolve during execution rather than being fixed at deployment.That expanding surface is exactly what a handful of runtime threats exploit.Goal drift happens when an agent begins with a legitimate objective but gradually deviates from the user’s original intent while attempting to optimize outcomes. An agent tasked with preparing a customer report, for instance, might autonomously retrieve unrelated confidential information because it incorrectly determines that additional context would improve the response.Excessive tool invocation is what happens when autonomous agents with access to numerous enterprise tools call unnecessary APIs, modify configurations, access sensitive repositories, or perform administrative actions simply because the model believes those actions are useful, absent any runtime controls to stop it.Memory poisoning exploits the persistent memory that improves personalization: Attackers can intentionally insert misleading instructions into long-term memory or retrieval systems, causing future decisions to be influenced by malicious or outdated information.Context manipulation takes advantage of how heavily LLMs depend on context: If attackers influence retrieved documents, system prompts, conversation history, or external data sources, they can indirectly steer autonomous behavior without ever compromising the underlying model. MITRE’s ATLAS framework catalogs this kind of adversarial behavior against AI systems in useful detail.Multi-agent amplification emerges as organizations deploy specialized AI agents that collaborate: If one agent behaves incorrectly, downstream agents may trust and amplify those actions, creating cascading failures across enterprise workflows.Introducing runtime trustRuntime trust extends security beyond authentication by continuously validating AI behavior throughout execution. Rather than assuming authenticated agents remain trustworthy indefinitely, it continuously evaluates whether autonomous decisions remain aligned with organizational policy. A runtime trust architecture rests on several complementary capabilities.Intent validation evaluates, before executing sensitive actions, whether proposed behavior still matches the user’s original objective: Is this action necessary? Is it expected? Does it exceed the requested scope? Would a reasonable human perform the same action?Behavioral monitoring observes tool usage, API activity, reasoning patterns, execution frequency, delegated actions, and abnormal workflows, so unexpected behavior becomes immediately visible rather than remaining hidden inside model reasoning.Policy enforcement means enterprise policies govern what AI agents can do, not merely what they can access — blocking financial transactions above approval thresholds, preventing privilege modifications, restricting administrative operations, limiting sensitive data retrieval, and requiring approval for high-risk actions. These controls function much like application firewalls for autonomous decision-making.Least-privilege execution means AI agents receive only the capabilities necessary for the current task. Instead of granting permanent access to dozens of enterprise tools, organizations should dynamically issue short-lived permissions based on runtime context, an approach that OWASP’s guidance for agentic applications increasingly emphasizes (OWASP GenAI Security Project).Human oversight recognizes that not every decision should be autonomous — high-impact operations, including financial approvals, identity changes, regulatory actions, or customer-impacting decisions, should require explicit human confirmation before execution.Protecting the enterprise AI ecosystemRuntime trust also extends beyond individual agents. As MCP adoption accelerates, enterprises should verify trusted servers, authenticated tools, approved capabilities, monitored interactions, and policy enforcement. RAG knowledge repositories require document integrity, source validation, access control, retrieval auditing, and poisoning detection. Persistent AI memory should implement lifecycle management, expiration policies, integrity verification, access logging, and sensitive data protection.Building operational visibilityOne of the biggest challenges in enterprise AI is observability. Security teams need visibility into why an agent selected particular tools, which data influenced its decisions, how it reached its conclusions, what actions it executed, whether policies were triggered, and which safeguards prevented unsafe behavior. Runtime logging, audit trails, and behavioral analytics are becoming essential components of enterprise AI operations, not optional add-ons.A practical roadmapOrganizations do not need to rebuild existing security programs. Instead, they should extend them by incorporating runtime trust into existing governance processes. Practical first steps include inventorying AI agents and their capabilities, applying least-privilege access to tools and APIs, classifying high-risk autonomous actions, implementing runtime policy enforcement, monitoring behavioral anomalies continuously, protecting memory and RAG data sources, requiring human approval for critical operations, and integrating AI runtime telemetry into existing SOC workflows.Looking aheadEnterprise AI will continue evolving toward increasingly autonomous systems capable of collaborating, planning, and executing complex business processes. Security strategies must evolve alongside them. The question is no longer whether an AI agent successfully authenticated. The more important question is whether it continues to behave safely throughout its entire lifecycle. Organizations that adopt continuous runtime governance today will be significantly better positioned to deploy autonomous AI responsibly, reduce operational risk, and build the confidence necessary for large-scale enterprise AI adoption.The future of AI security will not be defined solely by stronger models or better authentication. It will be defined by our ability to establish, measure, and continuously verify trust while intelligent systems are making decisions in real time.Ravindra Annam is a cyber security architect.

After spending the last several years in one of America’s least trusted, most opaque industries, the most competitive advantage is trust.

Operation Economic Outcast’s first bank action is a proposed rule on an Egyptian lender’s foreign unit. The campaign auctions dollar access.

U.S. stock-index futures were little changed on Sunday, as investors ponder the likelihood of a fresh interest-rate hike and look ahead to fresh labor data and tech earnings this week.

Marvell Technology (MRVL) stock fell about 10% on August 28, closing at $216.62, even after the chipmaker reported fiscal second-quarter results that came in ahead of expectations.

The drop looked harsh for a company that raised its outlook. But the selling had more to do with how high expectations had climbed than with anything Marvell did wrong.

That is the setup Morgan Stanley leaned into. 

The firm raised its price target while keeping a neutral rating, and it spent most of its note explaining why the data center business is the part investors should watch.

Why Morgan Stanley sees good news inside Marvell’s data center numbers

Morgan Stanley analyst Joseph Moore raised his Marvell price target to $246 from $224, keeping an Equal-weight rating, in a Morgan Stanley research note shared with me.

Moore has covered semiconductors at Morgan Stanley for years, and is one of the most closely followed chip analysts on Wall Street, so his read on Marvell carries weight even when he stays neutral.

The headline change was simple. Marvell now expects its data center segment to grow about 60% in calendar 2027, up from a prior estimate of about 50%.

More AI Stocks:

BMO starts Micron stock at Outperform with a $1,300 target

BMO starts Broadcom at Outperform with $455 target

Citi resets Marvell stock price target ahead of earnings

That single revision drives roughly 10% of added earnings power, according to the note.

Data center revenue already makes up close to 79% of Marvell’s total sales, so faster growth there moves the whole company.

Data center capital spending at companies like Alphabet’s Google and Amazon feeds Marvell directly, which is why analysts track those budgets so closely.

Marvell’s data center chips power the servers behind AI workloadsSundry Photography / Getty Images

What Marvell actually reported in its fiscal second quarter

Marvell delivered fiscal second-quarter revenue of $2.74 billion, up 13% from the prior quarter and 37% from a year earlier, according to the note and Marvell‘s investor relations page.

Non-GAAP earnings came in at $0.94 a share, slightly above both Morgan Stanley’s estimate and the broader Wall Street consensus.

Data center revenue hit $2.17 billion, up 18% sequentially and 46% year over year.

Here is the plain version of what those numbers mean:

The quarter at a glance

Revenue beat expectations, with data center growth the main engine.

Earnings edged past estimates at $0.94 a share.

Guidance for the October quarter points to about $3.15 billion in revenue, up roughly 52% from a year earlier.

Non-GAAP figures strip out items like stock-based compensation to show underlying operations. For Marvell, that gap matters, and we will come back to it.

Why the stock dropped even though the news was good

The confusing part for investors is why a beat-and-raise quarter sent the stock lower.

The answer sits in the Google deal announced days before earnings.

Marvell disclosed its expanded custom-chip partnership with Google on August 19.

Related: Broadcom stands to gain from new cloud deal

The agreement lets Google buy up to 58.97 million Marvell shares at $206.58 each, worth as much as $12.2 billion.

Investors expected that deal to add a fresh layer of growth on top of guidance.

Instead, Marvell’s management made it clear that the revenue from Google through fiscal 2028 was already priced into earlier forecasts.

So the number that excited traders turned out to be already counted, and the stock gave back its recent gains.

How Marvell is changing what it sells

The main focus here is a shift in Marvell’s business model, and Morgan Stanley likes the direction.

For years, Marvell’s growth leaned heavily on large custom chips known as ASICs, which are processors built for a single customer’s exact needs.

Competing head-to-head with Nvidia (NVDA) in that space is difficult, and big custom wins do not always land.

Now Marvell is spreading its bets across a wider set of products, which reduces the risk of a deal falling through.

That mix includes:

Optical DSPs, chips that keep data moving quickly across networks.

Data center switching and interconnects.

A group of smaller “attach” products such as network cards and storage controllers.

Marvell’s interconnect business is expected to grow more than 70% this fiscal year, well above its earlier target.

A more spread-out product line gives Marvell steadier growth, which is exactly what long-term investors tend to reward.

What still has to happen before the stock catches up

Morgan Stanley stayed Equal-weight for a reason, and it comes down to price.

Marvell trades at a rich valuation, and the firm noted that heavy stock-based compensation cuts into the case.

For the shares to work from here, a few things need to line up:

Estimate revisions need to keep moving higher, not just hold steady.

Marvell has to show it is winning more networking share over time.

Custom and attach revenue must grow faster than the current model assumes.

There are real risks as well. 

A smaller-than-expected AI market or a slowdown in enterprise data center and networking demand could pressure results, according to the note.

Marvell’s October investor day is the next event that could shift the story, and Morgan Stanley said it would turn more positive on any meaningful pullback.

How Marvell’s run stacks up against the market

Marvell has been one of 2026’s standout chip names, up roughly 195% year to date before earnings.

That run outpaced the broader market by a wide margin, with the S&P 500 posting far smaller gains over the same stretch.

The pullback after earnings trimmed some of that lead, though the stock still sits well above where it started the year.

The bottom line for investors

Marvell gave investors a strong quarter and a better long-term outlook, and Morgan Stanley responded by lifting its target to $246.

The stock fell anyway because the exciting Google numbers were already counted, not because the business weakened.

For investors, the useful signal is the shift toward a broader, steadier product mix, which lowers the odds that one lost deal sets the company back.

The catch is valuation. At current prices, much of the good news is already reflected, so the bigger opportunity may come if the stock cools off before the October investor day.

Marvell looks like a healthy business trading at a demanding price, and patience could matter more than speed here.

Related: Citi renews Nvidia stock forecast ahead of earnings