Welcome to this edition of our Weekly Analyst Insights roundup, which features key insights that our analysts have developed based on the past week’s events.
Context, context, context. For me, that’s often what separates wannabe tech analysis from the real deal. I’m thinking about this now because our expert in smart wearable devices, Anshel Sag, published two detailed — and context-rich — articles last week about important recent entrants into that market: the Samsung Galaxy XR headset and the Meta Ray-Ban Display glasses.
Two views of Moor Insights & Strategy VP and principal analyst Anshel Sag: On the left, wearing Meta Ray-Ban Display glasses to moderate a panel at this year’s MIT Reality Hack event; on the right, testing the thermal properties of the Samsung Galaxy XR headset. (Credit: Anshel Sag)
On one level, these pieces function as product reviews — and Anshel was a professional device reviewer before he ever became an analyst, so you’re in good hands on that score. But Anshel’s reviews also bring to bear his knowledge of previous device generations, industry history, competitive dynamics, and supply-chain relationships affecting the evolution of these devices. It’s remarkable how often I can read one of his fresh reviews, click through to a linked piece of coverage he wrote in some previous year, and come away thinking, “Yep, he saw this coming, and now it’s here.”
This week, Jason and I will be at IBM Think in Boston. Mike is attending the Oracle Analytics Summit in Redwood Shores, California, and Mel will be at ServiceNow Knowledge 2026 in Las Vegas. The spring season of client and vendor events is in full swing, and we’re on the road a lot over the coming months — including at SAP Sapphire, Zendesk Relate, and Dell Technologies World this month. If you’ll be attending any of the same events or see that we’ll be in your city, please reach out.
Last week, Moor Insights & Strategy analysts’ perspectives appeared across a broad set of business and technology outlets, including TechTarget, Data Center Knowledge, CNBC, Business Insider, Wired, Benzinga, CoStar, TechNewsWorld, Tech Insider, Machine Brief, and IT Brief Asia. Media coverage focused on accelerating AI infrastructure demand and hyperscaler spending, the growing influence of custom silicon from NVIDIA, Google, and Amazon, and emerging innovation in optical interconnects and AI cloud platforms. Analysts also weighed in on enterprise AI momentum as reflected in Microsoft and Intel earnings, the expansion of agentic AI development platforms, evolving storage and valuation dynamics, and early signals of AI-native devices and next-generation gaming experiences.
I also appeared on CNBC to discuss Microsoft’s earnings, strong Copilot metrics, and broader hyperscaler earnings trends. Check out the full list of citations here.
Our MI&S team also published 20 deliverables — 1 Research Paper, 4 Research Notes, 3 Analyst Insight, and 12 Podcasts.
Check out this week’s Analyst Insights roundup for more from the MI&S team, including what’s top of mind for each analyst, our thoughts on vendor announcements, press quotes, and more.
Have a great week!
Patrick Moorhead
What’s Next With AWS?
Last week I attended the What’s Next With AWS event in San Francisco. The event produced three meaningful announcements — an OpenAI partnership extension, a significantly updated Amazon Q, and a revamped Amazon Connect portfolio — and my read is that AWS pushed these into the spotlight deliberately to avoid having them buried at the much larger AWS Summit event in June.
New Alliance With OpenAI — AWS Extends Its Lead in Model Choice
AWS Bedrock can now serve OpenAI’s commercial models. Until now, AWS could only serve OpenAI’s open-weight models, so this extends its lead in model choice across the major hyperscalers. AWS is also offering a new managed OpenAI agent service built on Bedrock Agent Core components. Two things stand out here. First, AWS has offered managed services before — SageMaker and EKS being the most prominent — but this is the first time it has extended the managed model to third-party components. Second, OpenAI is currently the only agent harness offered as a fully managed service. You can use other harnesses with Agent Core, including AWS’s own Strands harness, but you don’t get automatic scaling and updating capabilities without the managed tier. AWS and OpenAI also signaled future technical collaboration on models, but offered no specifics. We’ll know more at re:Invent in December.
Amazon Quick Unifies Multiple Products and Rallies Around Cowork Apps
Amazon has had many “Q”- or “Quick”-branded user products over the years, but this week we saw many of the product capabilities rolled into a new product called Amazon Quick. In reality, this is a new product focused on a new category, rather than a product consolidation or packaging exercise.I would categorize Amazon Quick as an enterprise-grade Cowork-style application. I’m calling this broader trend the “Claudification of AI”; both Google and Microsoft have similar offerings, and that is a good thing.
These assistive tools have taken much of what worked in agentic IDEs and made it available to line-of-business users. Amazon’s specific angle is an embedded knowledge graph to better manage user context — conceptually similar to using a tool such as Obsidian as an AI second brain. (My colleague Mike Leone is a strong proponent of that approach.) The implications are interesting: Add a local LLM to Amazon Q and you could have an AI client with reach comparable to Apple Intelligence. When I raised that possibility with the AWS team, they declined to comment on future developments given how new the product is — a reasonable response, but one that leaves the broader scope of the Q platform genuinely open. Worth watching closely.
Amazon Connect Becomes an Agent Suite — Sort Of
AWS also revamped its Amazon Connect pre-packaged agent offering. Connect started as an agentic customer service application based on Amazon’s own internal support infrastructure. It has since expanded into a full suite: The product formerly known as Connect is now Amazon Connect Customer, joined by Connect Talent (high-scale recruitment and hiring), Connect Health (clinical care delivery), and Connect Decisions (procurement).
One important distinction: these are not full SaaS solutions. They are designed to integrate with or sit on top of existing SaaS applications. For customers with homegrown systems or entrenched SaaS stacks, that model may be a fit. But there is a real downside around integration costs and compliance complexity that enterprises should assess carefully before committing. This reinforces a broader point worth repeating: SaaS still has a structural role to play in the agentic future. Agents alone cannot replace the business logic and underlying architecture that makes SaaS solutions valuable to enterprises. The market for agent overlays is real, but enterprises need to be clear-eyed that they are acquiring a component of a solution, not a replacement for one.
Microsoft Agent 365 — Right Instinct, Need to Learn More
Last week Microsoft also announced the availability of Microsoft Agent 365, and I expect it to be a centerpiece at Microsoft Build this spring. Agent 365 offers a similar set of capabilities to what we saw from Google’s Enterprise Agent Platform, but with one key unresolved question: What exactly is the relationship between Agent 365 and Azure AI Foundry? Google was explicit that its platform represented the next generation of the Vertex development stack. Microsoft has not offered that same architectural clarity, and enterprise customers will need that before they can make confident build decisions. Agent 365 appears well-positioned to manage and govern agents across the enterprise — particularly those built using Microsoft’s own tooling, including Copilot Studio — but the boundary between Agent 365 and Azure AI Foundry needs to be defined. My expectation is that Microsoft Build will fill in that picture. I’ll have more analysis once those details are on the table.
What’s Next With AWS?
Last week I attended the What’s Next With AWS event in San Francisco. The event produced three meaningful announcements — an OpenAI partnership extension, a significantly updated Amazon Q, and a revamped Amazon Connect portfolio — and my read is that AWS pushed these into the spotlight deliberately to avoid having them buried at the much larger AWS Summit event in June.
New Alliance With OpenAI — AWS Extends Its Lead in Model Choice
AWS Bedrock can now serve OpenAI’s commercial models. Until now, AWS could only serve OpenAI’s open-weight models, so this extends its lead in model choice across the major hyperscalers. AWS is also offering a new managed OpenAI agent service built on Bedrock Agent Core components. Two things stand out here. First, AWS has offered managed services before — SageMaker and EKS being the most prominent — but this is the first time it has extended the managed model to third-party components. Second, OpenAI is currently the only agent harness offered as a fully managed service. You can use other harnesses with Agent Core, including AWS’s own Strands harness, but you don’t get automatic scaling and updating capabilities without the managed tier. AWS and OpenAI also signaled future technical collaboration on models, but offered no specifics. We’ll know more at re:Invent in December.
Amazon Quick Unifies Multiple Products and Rallies Around Cowork Apps
Amazon has had many “Q”- or “Quick”-branded user products over the years, but this week we saw many of the product capabilities rolled into a new product called Amazon Quick. In reality, this is a new product focused on a new category, rather than a product consolidation or packaging exercise.I would categorize Amazon Quick as an enterprise-grade Cowork-style application. I’m calling this broader trend the “Claudification of AI”; both Google and Microsoft have similar offerings, and that is a good thing.
These assistive tools have taken much of what worked in agentic IDEs and made it available to line-of-business users. Amazon’s specific angle is an embedded knowledge graph to better manage user context — conceptually similar to using a tool such as Obsidian as an AI second brain. (My colleague Mike Leone is a strong proponent of that approach.) The implications are interesting: Add a local LLM to Amazon Q and you could have an AI client with reach comparable to Apple Intelligence. When I raised that possibility with the AWS team, they declined to comment on future developments given how new the product is — a reasonable response, but one that leaves the broader scope of the Q platform genuinely open. Worth watching closely.
Amazon Connect Becomes an Agent Suite — Sort Of
AWS also revamped its Amazon Connect pre-packaged agent offering. Connect started as an agentic customer service application based on Amazon’s own internal support infrastructure. It has since expanded into a full suite: The product formerly known as Connect is now Amazon Connect Customer, joined by Connect Talent (high-scale recruitment and hiring), Connect Health (clinical care delivery), and Connect Decisions (procurement).
One important distinction: these are not full SaaS solutions. They are designed to integrate with or sit on top of existing SaaS applications. For customers with homegrown systems or entrenched SaaS stacks, that model may be a fit. But there is a real downside around integration costs and compliance complexity that enterprises should assess carefully before committing. This reinforces a broader point worth repeating: SaaS still has a structural role to play in the agentic future. Agents alone cannot replace the business logic and underlying architecture that makes SaaS solutions valuable to enterprises. The market for agent overlays is real, but enterprises need to be clear-eyed that they are acquiring a component of a solution, not a replacement for one.
Microsoft Agent 365 — Right Instinct, Need to Learn More
Last week Microsoft also announced the availability of Microsoft Agent 365, and I expect it to be a centerpiece at Microsoft Build this spring. Agent 365 offers a similar set of capabilities to what we saw from Google’s Enterprise Agent Platform, but with one key unresolved question: What exactly is the relationship between Agent 365 and Azure AI Foundry? Google was explicit that its platform represented the next generation of the Vertex development stack. Microsoft has not offered that same architectural clarity, and enterprise customers will need that before they can make confident build decisions. Agent 365 appears well-positioned to manage and govern agents across the enterprise — particularly those built using Microsoft’s own tooling, including Copilot Studio — but the boundary between Agent 365 and Azure AI Foundry needs to be defined. My expectation is that Microsoft Build will fill in that picture. I’ll have more analysis once those details are on the table.
AI Navigator was the SAS Innovate centerpiece last week, and the whole story centered on governance, trust, and observability for AI in production. The product watches what your models are actually doing once they’re deployed, and it’s designed to catch drift, flag bias, and keep a record of every decision a model makes, so when a regulator asks six months later why a customer got denied, you have an answer. SAS is an analytics company, and watching it lead with that story instead of analytics says everything about where buyers are right now. Dataiku shipped Kiji Privacy Proxy a few days later, which strips sensitive customer information out of any data being sent to a model, so the model never sees the SSN or the birthdate, but the agent can still do its job. Two years ago this wasn’t even a product category. Now there’s a steady drumbeat of vendors shipping these capabilities, one after another, because they’ve realized enterprises won’t put AI into production if they can’t see what it’s doing. The whole spring conference season has been about this: trust, confidence, safety, auditability. That’s the buying conversation, and it’s only getting louder.
And all of that is just one half of the trust conversation. Sovereignty is the other half, and it keeps getting pushed to the back burner of every tech event while trust, governance, and observability keep swallowing the headlines. That doesn’t mean it has stopped moving: Every hyperscaler and its cousin kept shipping sovereign capacity last week. Microsoft launched Azure Local Sovereign Private Cloud at thousands-of-nodes scale, on top of last week’s $16 billion sovereign AI commitment in Australia. Google broke ground on a gigawatt AI hub in Vizag, India. Equinix has been stacking sovereign announcements all year and added SCX with SambaNova, a colo-based AI inference rollout where sovereignty shows up as a downstream use case. AWS keeps building out European Sovereign Cloud. Oracle’s been selling sovereign offerings across the U.K., Saudi Arabia, and France for two years. SAP lit up Trusted Cloud France with Thales. Honestly, it feels like most of these are glorified data residency stories with new branding. Nobody agrees on what “sovereign AI” even means anyway: run-on-prem … hyperscale-in-country … colo-in-the-middle … national-cloud partnership — those are just residency answers. Real sovereignty also covers who runs the model, who holds the keys, who controls the orchestration, and which government can show up with a warrant. Most vendors are skipping those layers entirely, or not externalizing it the way they should. I predict that whichever definition wins is going to lock customers in for a decade, the same way cloud-region picks did in 2014.
Cloud partnerships keep showing up as the dividing line in storage, and last week’s cleanest example is Panzura. It GA’d Nexus, which connects enterprise file shares into Microsoft 365 Copilot. It ships through Microsoft’s marketplace as a co-sell, so for Panzura, that’s the win. It puts the company in front of Copilot deployments it wouldn’t otherwise be invited to. For customers, the win is on access control. Nexus enforces the permissions that already exist on the file shares, so the LLM sees only what each user is already cleared to see. What makes Nexus interesting is the global file system piece. Keeping permissions intact in real time across distributed locations is harder than what SaaS file vendors have shipped, and the bigger storage names haven’t gone there. That’s the technical edge behind the Microsoft co-sell.
The partnership wins are real, but the hardware side of storage is a different story this week. Everpure CEO Charles Giancarlo’s open letter is the kind of note you almost never see from a public-company CEO. He told customers straight up that the 70% year-to-date price hikes aren’t going away, and that quote windows have compressed from 90 days to 30. That’s the part of the AI build-out that nobody on a vendor stage wants to talk about. Component costs are up 300 to 900%, and storage CEOs are publicly committing to not profiteer through it. Real talk like Giancarlo’s from someone running a public company is rare. If you’re sourcing storage right now, lock in pricing earlier than you think you need to.
Adobe closed its acquisition of Semrush, a tool I’ve used for years in marketing and competitive analysis. This brings Semrush’s vast keyword and backlink data into real-time generative engine optimization (GEO) within Adobe Experience Manager and Analytics workflows, just as agentic search gains steam.
Based on chats at the Adobe Summit a couple of weeks ago, Adobe plans to embed discoverability deeply into CX orchestration. Agentic attribution that ties AI/GEO visibility to enterprise results could support MarTech pricing premiums.
Execution will matter against rivals like Salesforce. I’m watching for early Experience Cloud integrations and client metrics showing brand reach across LLMs and search engines.
Quick thoughts on AWS “What’s Next?” event: Last week, I attended the What’s Next With AWS? event in San Francisco, where the company presented Amazon Quick as more of a coworker extension than the typical copilot. It leverages your knowledge graph, workflows, meeting notes, and preferences to amplify what you already do. Early demos impressed me, though I want hands-on time to really put it through its paces.
The company also unveiled Connect Talent, a 24/7 automated hiring platform for massive seasonal scaling. Amazon, which ramps up to 250,000 hires per week during peak times, is Customer Zero. Alongside it came “humorphism,” a design approach to make AI feel more natural in interactions. This could ease the unease around voice AI, moving past basic chat.
I think Connect Talent has clear potential to grow into something bigger. Whether this remains an AWS application or is expanded through partnerships, the company could build a massive pipeline of qualified workers across multiple sectors. For workers, this could become a “common app” of sorts (like the ones college applicants are so familiar with), allowing them to apply for or be pre-screened for multiple positions. For companies, this could create a talent bench that further shortens time to hire and onboarding. I’ll be watching to see how AWS takes this to the next level, but the use cases are promising.
Overall, AWS’s shift toward full enterprise apps marks a pivot from pure infrastructure to end-to-end solutions. This may accelerate agentic AI adoption in HR, supply chain, and CX workflows, pressuring CRM, ITSM, and HCM incumbents via native Quick agents and simpler integrations.
Box Automate is now generally available. The no-code workflow capability now supports content-driven processes such as contract approvals, onboarding document routing, and compliance reviews across Business plans and above, with full agentic features for Enterprise Advanced customers. This could include scenarios such as automatically routing a sales contract for legal review upon upload, or triggering a compliance checklist when sensitive documents are detected. Box enables file-event triggers for approvals, tasks, and notifications, integrated with Box AI and metadata — which could, for customers, reduce reliance on external tools.
More platforms are building automation directly into where content lives, and Box is making clear moves in that direction. For enterprise teams that already use tools such as Zapier or Power Automate for broader orchestration, Box Automate could be compelling enough to consolidate, or it could become a specialized layer alongside existing tools.
Half the enterprise data leaders I talk to are doing the math on whether they really need a separate vector database when their existing data platform can do it. Qdrant’s announcement last week was built to make that math harder for the existing platforms to win. The company added GPU-powered index builds for speed, multi-zone replication for uptime, and audit logs on every operation for compliance. Through 2023 and 2024, the question was whether vector databases even worked for AI. The 2026 question is whether they can stay up, log every retrieval, and rebuild indexes fast enough to keep agents fed with current data. Hyperscalers and lakehouse vendors are baking vector search into platforms customers already use, and the “Why use a separate database when mine already does it?” pitch is getting harder to fight. What Qdrant shipped last week opens up the set of enterprise buyers that were off-limits to them before. The customers who couldn’t take vector search to production because of uptime, compliance, or indexing gaps now have a Qdrant they can actually deploy. That’s a meaningful expansion of where Qdrant can compete in 2026.
A lot of AI deployments I see can’t go to a hyperscaler, can’t assume a steady internet connection, or can’t leave the building. That’s the part of the market Actian launched VectorAI DB into last week. It’s a portable vector database built to run at the edge, in regulated environments, and in disconnected workloads, all the way down to devices as small as an NVIDIA Jetson or a Raspberry Pi. Manufacturing floors, hospital datacenters, defense installations, and retailers running AI inference inside the store all fit this profile. None of them get served well by the cloud-native vector vendors. Actian’s bet is that this segment is bigger than it gets credit for and growing fast as enterprises figure out that not every AI workload should leave their walls. The harder lift is awareness. Actian sits under HCLSoftware, which sells across a much broader software portfolio than focused enterprise data. Most buyers default to a shortlist that doesn’t include Actian, and the edge-and-regulated buyers tend to sit in different procurement teams than the cloud-native vendors are calling on. Even so, Actian has earned a spot on my watchlist for the rest of 2026.
Dreame NEXT
Dreame (pronounced “dreamy”), a fast-growing Chinese consumer brand, has been flying under the radar in the U.S. for the past five years. The company entered the market with consumer products such as robot vacuums and personal care devices, available through online retailers. While it has a larger, more diverse product footprint in Europe and Asia, its presence in North America has been limited — until now.
Dreame held its first major U.S. launch event April 27–30, 2026, at the Palace of Fine Arts in San Francisco. Dreame NEXT served as the official kickoff for a significantly expanded U.S. product line and brand identity. The company unveiled an impressive array of innovations, including a rocket-powered car that, it says, can go from 0 to 60 mph in 0.9 seconds. I’ll leave it to the automotive press to reality-check that claim while we turn our attention to the company’s product strategies.
Best known in the U.S. for robot vacuums, Dreame has chosen to stake a claim in the future of smart consumer products rather than develop incremental features to compete against established companies. The company defines “smart” as robotic — a sense/think/act loop playing out simultaneously across many product categories, including large and small appliances, lawn and garden, personal care, entirely new robotic embodiments, and Matter-based smart home devices.
Looking deeper, two strategic pillars support the company’s quest for explosive growth in established consumer markets. (In addition to the rocket car, of course.)
Pillar 1: Physical Action
Dreame’s AI focus is motion and physical AI — sense/think/act loops applied to everyday consumer use cases. Most competitors focus AI development on sense and think — cameras, sensors, models, inference, connectivity. Dreame has chosen the action path, and it is investing accordingly. Consequently, the company outspends most of its peers on research and development. According to a recent company blog post, “R&D personnel comprise approximately 70% of the company’s workforce and R&D expenditure accounts for over 7% of annual revenue.” A large and rapidly growing patent portfolio supports this claim. As of December 2025, Dreame had filed more than 10,000 patents worldwide (with 3,000 granted).
My favorite data point for Dreame’s R&D efficacy comes in motor technology. Dreame holds Frost & Sullivan’s world-first certification for a 200,000 RPM digital motor and mass-produces it to achieve 160,000 RPM. At NEXT, the company introduced a 250,000 RPM magnetic-levitation motor in the XT Combo vacuum. A solid foundation of motor IP now drives vacuums, hair dryers, refrigerator compressors, and automotive applications. Clearly, this company is very serious about motion leadership.
Sensing follows a similar pattern. Dreame builds the DHX1 lidar in-house. A 2,160-line unit debuted at NEXT for the vehicle program. The N1 Refrigerator concept pairs an 8 MP binocular camera with a 32-channel hyperspectral (I’d call it “multispectral”) sensor operating in the 900–1700 nanometer band. Hyperspectral imaging is typically found in laboratory hardware; Dreame aims to put it in a kitchen appliance to identify and evaluate food items.
My take: If a company is just trying to beat incumbents at their own game, none of this profile makes sense. Dreame has decided to play a new game: physical capability. The R&D profile, the patent depth, and the in-house core components reflect that strategic choice.
Pillar 2: Robotics-First Product Design
Conventional appliance development focuses on incremental improvements within established product categories. Robotics-first development takes those product categories to the next level. Dreame appears to apply the robotics approach across every target category.
Here’s an example. The Cyber10 Ultra robot vacuum, shipping later in 2026, features a four-jointed arm with 5 degrees of freedom and a 500g lift capacity. Instead of navigating around clutter, this robot moves clutter. The closest competing product is Roborock’s Saros Z70 (300g lift, shorter reach, no tool-swapping).
Manipulation is not a vacuum-only feature. Dreame announced what it calls the world’s first dual-robotic-arm air conditioner, featuring two articulated arms that direct airflow into independently zoned regions of a room. The same arm technology appears in range hoods, steam ovens, and dishwashers. The Z1 Laundry Robot concept uses a multi-joint arm to transfer clothing between cycles autonomously. The N1 Refrigerator concept sorts groceries into compartments with an internal arm.
Dreame first introduced bionic robotic arm technology in 2023. In three years, it has crossed into climate, laundry, kitchen, and mobility. That’s impressive velocity, plus it’s structural evidence that a robotics-first design methodology is center stage — even if some of the use cases might be too advanced for many consumers.
Pillar 3: Matter
Here’s another example of Dreame’s R&D-driven, over-the-horizon product design strategy. The company uses Matter for smart home connectivity, signaling a move away from walled-garden isolation to unification around Matter, Thread, and Aliro standards.
Dreame is integrating Matter across much of its 2026 lineup, ensuring that connected products are interoperable across multiple smart home platforms, including Apple, Google, Amazon, and Samsung. Key examples include:
My take: The rocket car isn’t a category-redefining move. It’s a signal that Dreame is willing and able to take risks. However, bold product concepts can blur the line between technology theatrics and technical progress. So, as an analyst, I’ll focus on tangible progress that transforms traditional appliances from observers to actors — for instance, by adding physical manipulation and actuation. If that reframe lands with consumers, every white-goods incumbent competing only on incremental features and energy ratings has a strategic problem.
Dreame’s strategy is to turn a solid foundation of advanced robotics technologies into a true inflection point across multiple consumer product lines, and I think it has a good shot. However, execution risk is real, and consumers can be tough customers. The question for the next five years is whether consumers regard new robotics-first products as compelling, useful, reliable, and worth the money.
T-Mobile for Business has expanded its enterprise portfolio with the launch of SuperBroadband, a unique, turnkey-managed internet solution that combines the carrier’s 5G fixed wireless access (FWA) with SpaceX’s Starlink satellite connectivity. Designed to simplify business networking and eliminate costly downtime, SuperBroadband functions as either a seamless failover or a dynamic load-balancing solution, boasting an uptime of 99.99%. What makes this offering unique is that T-Mobile is the first major Tier-1 carrier to natively integrate low earth orbit (LEO) satellite into its own managed networking stack, ensuring that connectivity reaches literally every zip code in the United States. For a flat $250 monthly fee, businesses receive a package that includes all hardware, powered initially by Ericsson and soon Inseego routers alongside a Starlink antenna, plus installation, configuration, and ongoing management via the T-Platform. Ultimately, this single-vendor approach promises to provide significant value by allowing distributed enterprises, such as early adopters Columbia Sportswear and Shell, to replace dozens of fragmented regional ISPs with a highly reliable, simplified, and unified broadband strategy.
Nokia announced that it would sell its FWA business to Inseego and become an investor in the company, which is a major player in the CPE (customer premises equipment) space. I believe that this is a consequence of Nokia wanting to focus more on its core businesses and give its FWA division a chance to compete with the likes of Cradlepoint, which is now part of Ericsson. Seeing as Ericsson is one of Nokia’s biggest competitors, it does make sense that Nokia would still want to see the division be successful, even if it’s within a standalone company like Inseego. As mentioned in the item just above, Inseego is also one of the launch partners for T-Mobile’s new FWA-based SuperBroadband product, which also includes Ericsson’s Cradlepoint.
I attended the T-Mobile for Business SuperBroadband launch last week. This is a managed service that combines T-Mobile’s nationwide 5G network with Starlink satellite connectivity. The offering provides dual network paths with a stated uptime target above 99.99% and coverage across all U.S. ZIP codes, including remote areas.
The company positions the service as an alternative to multi-ISP configurations, which often require multiple separate contracts and manual failover. By integrating terrestrial and satellite connectivity, T-Mobile aims to simplify resilience for distributed enterprises.
The emphasis on uptime reflects the reality of real-time, transaction-driven environments. In a prior role where I supported a large-scale, distributed payments operation, even a 30-second outage could halt transactions and cause immediate revenue and customer impact. That experience shows me where a unified, resilient connectivity model like this should resonate, particularly across retail and other high-volume transaction settings.
Quantum Leap or Quantum Lag
H.R. 8462, the National Quantum Initiative Reauthorization Act (NQIRA), is an important piece of legislation introduced by Representative Randy Weber (R-Texas). Since the bill was cleared by the House Committee on Science, Space, and Technology on April 29, it is now awaiting a full vote by the House of Representatives. The original National Quantum Initiative (NQI) was approved in 2018 and focused on basic laboratory science and issues such as interagency coordination. In 2019, I wrote a Forbes article entitled “Quantum USA Vs. Quantum China: The World’s Most Important Technology Race.” For H.R. 8462 to be effective, it must focus on both applied research and commercialization. Its primary science goals should still include maintaining a lead in quantum information science over China.
Workforce development is essential to ensure a robust talent pool. That includes government funding to strengthen and expand National Science Foundation fellowships and traineeships to build a much-needed pipeline of quantum-trained workers. The United States must also increase the availability of U.S.-made hardware needed for quantum components used in our infrastructure and supply chains. The future creation of domestic foundries is included in critical infrastructure roadmaps.
Challenges to Passing H.R. 8462
Even though there is a broad bipartisan consensus that we must be the winners in quantum, the bill faces several hurdles:
Risks of Failure If H.R. 8462 Doesn’t Pass
Proponents of H.R. 8462 argue that failing to reauthorize and modernize this initiative would lead to several critical problems:
National security vulnerabilities: Quantum sensing is vital for GPS-independent navigation, submarine detection, and ultra-sensitive telecommunication applications. There have been significant breakthroughs in this important research, but lack of funding could leave the U.S. military at a disadvantage in future “dark” or electronically jammed environments.
Lots of big tech companies announced earnings last week, including Apple, Google, Meta, Microsoft, and Qualcomm. Apple showed significant growth in China after taking a hit in previous quarters and saw Mac growth improve alongside other businesses. Google also showed strength in its earnings, with significant growth in Google Cloud and an overall payoff on its efforts in AI with Gemini. Google even saw 19% growth in search, which many people have written off as a dying business because of AI. Meta reported very strong advertising revenue growth, but also increased its expectations for capex again, which impacted some of its outlook numbers (for which it got punished in the stock market). Qualcomm reported earnings that demonstrated a clear impact from the memory crunch as Chinese OEMs get hit the hardest. However, the company did also show major growth in its automotive business and announced a custom ASIC solution with a hyperscale partner. Qualcomm also seemingly indirectly confirmed that it was working with OpenAI on a smartphone. Last but not least, T-Mobile reported earnings with strong post-paid net adds and guided upward for the full year of 2026, exciting investors.
RingCentral has rolled out RCS Branded Messaging for verified texts showing logos and details inside native apps, plus Enterprise Branded Calling so outbound calls display company info. It has also expanded international SMS, extended AI Receptionist to cover SMS inboxes and queues, and added a Customer Engagement Bundle with Operator Connect for Microsoft Teams.
These additions address real problems for companies: Spam filters can block legitimate messages, and siloed channels can drop context. Branded calls and texts should get more pickups, and global SMS could smooth out delivery overseas.
Beyond that, AI handling routine interactions via voice and text, plus Teams embedding, may reduce missed chats and wait times. This aligns with the shift to all-in-one setups, where AI conversations feel more standard in support. In my view, RingCentral has done a good job of staying ahead by innovating on AI that integrates natively into familiar tools like Microsoft Teams, simplifying adoption for users while enhancing CX through contextual, real-time automation.
Six Five (Patrick Moorhead)
GoodData / AI, Agent Builder / Mike Leone / Tech Target
GoodData joins agentic AI development mix with Agent Builder
Hyperscaler Earnings / Patrick Moorhead / DataCenter Knowledge
Analysis: Hyperscaler Earnings Show AI Demand Outrunning Infrastructure
Intel / Stock, Earnings / Patrick Moorhead / CNBC
Intel’s stock more than doubles in April for best month in chipmaker’s 55 years on Nasdaq
Kopin / AI Infrastructure / Matt Kimball / IT Brief Asia
Kopin & Fabric.AI back MicroLED optical links for AI
Microsoft / Earnings / Patrick Moorhead / Benzinga
Microsoft’s AI Engine Roars: Analysts See Real Demand, Not Just Hype
NVIDIA, Google, Amazon / AI Chips / Patrick Moorhead / Business Insider
Nvidia’s $4.9 trillion chip empire has a new problem: its biggest customers
NVIDIA, Google, Amazon / AI Chips / Patrick Moorhead / Machine Brief
Google and Amazon Eye AI Chip Supremacy: A New Era or Empty Ambition?
OpenAI / Smartphones / Anshel Sag / Tech News World
OpenAI Eyes AI Agent Phone, Kuo Says
Sony / Video Games, PS5 / Anshel Sag / Wired
Saros Shows Off the PS5’s DualSense Tricks
Tech Giants & Data Center Spending / Patrick Moorhead / CoStar
Tech giants defend AI spending, shell out billions more
Vast Data / AI Storage, Valuation / Patrick Moorhead / Tech Insider
Vast Data $1B Round at $30B: NVIDIA’s AI Storage Bet [2026]
Verda / AI Cloud / Matt Kimball / DataCenter Knowledge
Verda Bets $117M on AI Cloud as Workloads Fragment Across Platforms
Unless otherwise noted, our analysts will be attending the following events in person.
August events coming soon.
September events coming soon.
December events coming soon.
Founder, CEO and Chief Analyst | + posts
Patrick Moorhead is the founder, CEO, and chief analyst of Moor Insights & Strategy. His big-picture view of technology is grounded in more than 20 years as an executive leading strategy, product management, product marketing, and corporate marketing functions at NCR, AT&T, Compaq, and AMD. He has shared his expertise in areas from silicon to infrastructure to enterprise SaaS and everything in-between in thousands of national broadcast appearances (CNBC, Yahoo Finance), articles (Forbes, CIO), research-based analyses, and podcast episodes. Today, he has 100+ CXO-level advisory clients and is often ranked the #1 technology industry analyst by ARInsights.
Paul Smith-Goodson is the Moor Insights & Strategy Vice President and Principal Analyst for quantum computing and artificial intelligence. His early interest in quantum began while working on a joint AT&T and Bell Labs project and, during 360 overviews of Murray Hill advanced projects, Peter Shor provided an overview of his ground-breaking research in quantum error correction.
Jason Andersen is vice president and principal analyst covering application development platforms, technologies, and services. Jason brings over 25 years of experience in product management, product marketing, corporate strategy, sales, and business development at Red Hat, IBM, and Stratus to his work for MI&S and its advisory clients. Working both in the field and in the headquarters of some of the most innovative technology companies, Jason has a wealth of experience in building great products and driving their adoption across a broad spectrum of industries and use cases.
Senior Analyst-in-Residence | + posts
Bill Curtis is the Moor Insights & Strategy Analyst in Residence for large-scale Internet of Things systems. Bill helps enterprises design distributed solutions that integrate the full end-to-end IoT stack from real-world devices to analytics.
Matt Kimball is a Moor Insights & Strategy senior datacenter analyst covering servers and storage. Matt’s 25 plus years of real-world experience in high tech spans from hardware to software as a product manager, product marketer, engineer and enterprise IT practitioner. This experience has led to a firm conviction that the success of an offering lies, of course, in a profitable, unique and targeted offering, but most importantly in the ability to position and communicate it effectively to the target audience.
Mel Brue is vice president and principal analyst covering modern work and financial services. Mel has more than 25 years of real tech industry experience in marketing, business development, and communications across various disciplines, both in-house and at agencies, with companies ranging from start-ups to global brands. She has built a unique specialty working in technology and highly regulated spaces, such as mobile payments and finance, gaming, automotive, wine and spirits, and mobile content, ensuring initiatives address the needs of customers, employees, lobbyists and legislators, as well as shareholders.
Mike Leone is a principal analyst at Moor Insights & Strategy covering data platforms and analytics, data infrastructure and storage, and data governance and enterprise data strategy. He brings 15 years of analyst experience from his work at Enterprise Strategy Group, where he rose to practice director for data management, analytics, and AI. Mike's work is grounded in a strong technical and strategic foundation, including early roles in software and hardware engineering.
Anshel Sag is Moor Insights & Strategy’s in-house millennial with over 18 years of experience in the IT industry. Anshel has had extensive experience working with consumers and enterprises while interfacing with both B2B and B2C relationships, gaining empathy and understanding of what users really want. Some of his earliest experience goes back as far as his childhood when he started PC gaming at the ripe of old age of 5, building his first PC at 11, and learning his first programming languages at 13.
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。