We are tracking the trends that matter most for leaders planning the next 12 to 36 months. We organize trends into three categories based on their maturity and impact timeline:

Game Changers: Trends reshaping entire industries right now. These affect how businesses operate, how value is created, and how competitive advantage is built.

Foundational Breakthroughs: Scientific and engineering advances unlocking new possibilities. These create strategic optionality for the decade ahead.

Weak Signals: Early indicators of transformations that will dominate the late 2020s. These require positioning now, even if mainstream adoption is years away.

Today’s Foundational Breakthrough: air taxis, and what they could mean for how people and goods move, in cities and beyond.

Air taxis are closer than you think: What leaders should know

Electric air taxis and autonomous aircraft are moving from prototype to certification. These electric vertical takeoff and landing aircraft, or eVTOLs, take off and land like helicopters but are quieter, electric, and designed to be cheaper to operate. That makes them useful well beyond dense cities, from suburbs to regional trips to places with no easy road access.

For decades, the flying car lived in concept sketches and science fiction. Now aircraft are flying real routes between real cities, regulators are writing the rules, and vertiports are being built. The first routes are in cities because that is where demand and money are concentrated, but the technology isn’t limited to them.

Dubai is furthest along. Joby Aviation recently added three additional vertiports to its Dubai electric air taxi network, at Dubai Mall, Atlantis The Royal, and the American University of Dubai. They join the flagship hub beside Dubai International Airport, which received regulatory approval this summer.

In its November 4 shareholder letter, Joby said the second vertiport in the network is now complete and that piloted operational flight testing across the network will begin later this year, with passenger service to follow. Once running, the service will turn a 45-minute drive from the airport to Palm Jumeirah into a flight of roughly 10 to 12 minutes.

Who’s building advanced air mobility

Joby Aviation is closing in on certification. Founded in 2009 by JoeBen Bevirt on his ranch in the Santa Cruz Mountains, Joby spent its first decade developing its aircraft largely out of public view. It is still headquartered in Santa Cruz, with Toyota as a major investor and manufacturing partner. Joby is now in the fifth and final stage of FAA type certification, with flight testing by FAA pilots expected to begin in the first half of 2027. It already carries passengers today through Blade, the helicopter passenger business it acquired, which flew more than 44,000 people last quarter. That gives Joby a customer base and booking channel ready and waiting for its electric aircraft.

Alef has begun production of flying cars. The California startup began building its Model A in December 2025: a fully electric vehicle that drives on public roads and takes off vertically, with about 220 miles of driving range and 110 miles of flight range. Alef reports more than 3,600 preorders, and its first units will go to a small group of early customers, no more than 10, who will fly them under tightly controlled conditions. Alef had targeted the end of 2026 for those first deliveries, but in late October, CEO Jim Dukhovny said he now aims to deliver the first car in 2027.

Archer’s Midnight aircraft is completing key flight milestones. Midnight flew roughly 55 miles in 31 minutes on a piloted test flight in 2025, its longest at the time. This fall, Archer took the aircraft into real urban airspace with its “No Roads” tour, which culminated in piloted flights around San Francisco Bay, including an Oakland to Marina District hop of about 10 minutes, a drive that can top an hour in traffic. Archer has since unveiled planned air taxi networks for Austin, Houston, and Dallas-Fort Worth, and is preparing for early U.S. operations under the federal eVTOL Integration Pilot Program as it works through FAA type certification.

Skyports is building the Dubai vertiports Joby will fly from, with earlier projects in Singapore and Paris. Its flagship vertiport, the Dubai International Airport hub where Joby’s network starts, spans four floors and 3,100 square meters, with capacity for 42,000 aircraft movements and about 170,000 passengers a year. It also includes a multilevel car park and sits a short walk from the metro. Vertiports designed to plug directly into existing transit are what turn a novelty into a network.

Why it matters

Think of early rideshare: niche at first, then part of everyday life. When rideshare launched, it was a premium black car service for a narrow slice of riders. Within a decade, it changed how people get to the airport, how cities think about parking, and what consumers expect from getting around: tap a screen, get picked up, pay automatically.

Air mobility looks to follow a similar path. Early routes will be premium, limited, and concentrated in a handful of places with the political will to move first. Those places won’t all be big cities. Regions with long drives, difficult terrain, or island geography have a strong case too. The pieces are coming together quickly. Joby air taxi trips in Dubai are set to be bookable directly in the Uber app, and autonomy is advancing in parallel: in September, Joby flew a converted Cessna Caravan 3,199 miles across the United States on its autonomous system, with the onboard safety pilot never touching the controls.

For leaders, the implications reach well beyond aviation. Real estate developers are already treating vertiports as a premium amenity. Airports, hotels, and corporate campuses will compete on air connectivity. Business travel, logistics, and emergency response will each find their own use cases as costs come down and networks fill in. Medical transport and cargo to remote areas are likely to be among the earliest.

Organizations that start planning for a vertical layer of transportation now will be ready when it reaches their markets, whether those are urban, suburban, or rural.

This is part of our ongoing look at the trends shaping the next decade. Explore the full 12 Tech Trends to see what else is on our radar.

Skydriven Mobility: Cities prepare for air-first commuting was originally published in Megatrends by HP on Medium, where people are continuing the conversation by highlighting and responding to this story.

We’re taking a look at the trends that matter most for leaders navigating the next 12 to 36 months.

We organize trends into three categories based on their maturity and impact timeline:

Game Changers: Trends reshaping entire industries right now. These affect how businesses operate, how value is created, and how competitive advantage is built.

Foundational Breakthroughs: Scientific and engineering advances unlocking new possibilities. These create strategic optionality for the decade ahead.

Weak Signals: Early indicators of transformations that will dominate the late 2020s. These require positioning now, even if mainstream adoption is years away.

Today, we’re diving into one of the Foundational Breakthrough trends: Omnimodal Intelligence.

What is it?

Multimodal AI (text + image + audio) was only the beginning. Omnimodal systems can combine vision, language, spatial data, code, simulation, physics, and robotic action, enabling AI to understand not just our digital worlds but the physical world around us.

A good metaphor for this is the human brain. We don’t have one brain for seeing, one for hearing etc. We have one integrated brain that considers all our senses, all at once. This is the foundation for robotics, AR, autonomous systems, and digital environments that understand us as richly as we understand them.

Where we’re seeing it emerge

Every sense fused into physical action. Figure’s Helix 02 folds vision from head and palm cameras, tactile sensing fine enough to register forces as small as three grams, proprioception, and natural language into a single neural network that controls an entire humanoid robot.

Rather than bolting a locomotion controller onto a separate manipulation system, one model lets the robot walk, balance, and manipulate as a single continuous system, running entirely from onboard sensors with no human intervention. In one demonstration, it completed a continuous four-minute task made up of 61 separate actions, described as the longest-horizon autonomous task a humanoid has performed to date. It signals where omnimodal is heading: perception fused tightly enough that a machine can operate inside the physical world it senses.

One model that sees, listens, and moves a whole body. In July 2026, Google DeepMind introduced Gemini Robotics 2, a family of models that takes in live video, audio and language and turns them into action. Its reasoning model can talk with people, track its own progress on a task by watching it, and plan multi-step work, and in one demonstration it paused a humanoid when a person walked nearby and resumed only once the area was clear. The action model now directs an entire humanoid, not just its hands, and an on-device version runs on the robot’s own computer and can be adapted to a new robot with a few hours of training. Omnimodal perception is moving onto the machine itself, where the decisions happen.

3D worlds built for machines to train in. World Labs’ Marble generates persistent 3D worlds from text, images, video, or coarse layouts, and it does so as a multimodal world model rather than a single purpose image tool. What makes Marble a genuine signal of omnimodal progress is where its output goes next: exported as Gaussian splats or collider meshes, these AI generated environments are already being pulled directly into NVIDIA Isaac Sim to train robots, cutting scene setup from weeks to hours. Marble is a bridge between generative AI and the physical systems that have to operate in the real world.

A robot brain that imagines before it acts. NVIDIA’s next humanoid model, GR00T N2, previewed in March 2026, is built as a “world action model.” Rather than mapping what a robot sees straight to a movement, it first predicts how the scene will change, then plans the actions that get it to the goal. NVIDIA says this helps robots succeed at new tasks in new environments more than twice as often as leading vision-language-action models, and N2 currently ranks first on the MolmoSpaces and RoboArena benchmarks for generalist robot policies. It is slated to be available by the end of the year.

The world’s robots learn in are changing too. Cosmos 3, which NVIDIA released as an open model in June 2026, handles text, images, video and robot actions in one system, folding scene understanding, world generation and action simulation together. NVIDIA calls it the first open omni-model for physical AI reasoning and action. Together they show omnimodal perception moving from understanding a scene to controlling a body inside it.

Omnimodal perception as the eyes and ears of AI agents. Most agent systems today still stitch together separate models for vision, speech, and language, losing time and context every time information passes from one model to the next. NVIDIA’s Nemotron 3 Nano Omni, released in April 2026, folds text, image, audio, video, and document understanding into a single open model built for exactly this handoff problem, delivering up to 9x the throughput of comparable open omni models. It is built for computer use agents, document intelligence, and audio and video reasoning. Palantir and Foxconn are among the companies already adopting it, and others such as Oracle and Docusign are evaluating it, an early sign that omnimodal perception is moving from research benchmark to production agent stack.

Why it matters

For years, “multimodal” meant a model that could look at an image and describe it or transcribe audio into text. Omnimodal is a different ambition: systems that hold vision, sound, language, spatial structure, and physical dynamics in one coherent understanding, the way a person walking through a room does without thinking about it.

That shift changes what AI can do in the places where decisions actually get made:

Decisions made in context. An omnimodal system can read voice tone, facial cues, documents, dashboards, and live video at once, connecting a spike in sensor data to the line in a written report that explains it and picking up both what is said in a meeting and what is left implied. Leaders get fewer blind spots because the signals no longer sit in separate tools that never talk to each other.

Holistic diagnosis. In medicine, one model can weigh imaging, genomics, lab results, and a patient’s speech patterns together, flagging anomalies that surface across modalities rather than in any single test. That gives clinicians integrated insight instead of four disconnected readouts, pointing toward earlier detection and better outcomes.

Just-in-time operations. On the factory floor, omnimodal perception fuses visual inspection, IoT sensor streams, and maintenance logs into real-time anomaly detection across machines and whole facilities, with hands-free guidance delivered through voice and augmented visuals. The payoff is reduced downtime, safer environments, and predictive intervention before failure instead of after.

Immersive learning. In training and education, a system can learn from how a student speaks, writes, gestures, and interacts, then adapt instruction across AR/VR, video, and dialogue and drop the learner into experiential simulations. Because instruction meets each person where they actually learn, retention climbs and training moves faster.

The through line is the same in every case: when perception stops being fragmented, a system can act on the full picture instead of a slice. For enterprise leaders and founders building in this space, the strategic question is where perception stops being a bottleneck. Teams working on robotics, spatial computing, simulation, healthcare, industrial operations, or agentic AI are the ones who will feel this shift first, and the infrastructure choices they make now, which world models to build on, which perception stacks to bet on, will shape what they can ship over the next several years.

Omnimodal Intelligence: AI that perceives the entire world was originally published in Megatrends by HP on Medium, where people are continuing the conversation by highlighting and responding to this story.

We’re looking at the trends that matter most for leaders navigating the next 12 to 36 months.

We organize trends into three categories based on their maturity and impact timeline:

Game Changers — Trends reshaping entire industries right now. These affect how businesses operate, how value is created, and how competitive advantage is built.

Foundational Breakthroughs — Scientific and engineering advances unlocking new possibilities. These create strategic optionality for the decade ahead.

Weak Signals — Early indicators of transformations that will dominate the late 2020s. These require positioning now, even if mainstream adoption is years away.

Today, we’re diving into a Foundational Breakthrough: Fusion.

Fusion is the reaction that powers the sun: forcing hydrogen atoms together until they release enormous amounts of energy. Harnessing it here on Earth has been a scientific dream for decades, and the promise is hard to overstate. Fusion could deliver clean power that runs around the clock, pulls its fuel from seawater, emits no carbon, produces little long-term radioactive waste, and carries no meltdown risk. It’s the closest thing we have to a safe, effectively limitless energy source.

Recently, fusion is crossing from physics research into engineering reality. This year, Helion Energy became the first company in the world to secure the regulatory licenses needed to build and operate a fusion power plant, and it continues construction and manufacturing buildout at its Orion facility in Washington state. While investment in the fusion industry has skyrocketed over the past year, reaching $4.48 billion, as a result of the demands of AI data centers and rising global energy concerns.

Breakthrough momentum

• Helion cleared a historic regulatory milestone in June 2026, receiving the Radioactive Materials and Radioactive Air Emissions licenses from Washington’s Department of Health, the first such approvals granted to any fusion company. Assembly and office buildings at Orion are complete, generator building earthwork is underway, and Helion raised $465M in a Thrive Capital-led round that nearly tripled its valuation to $15.5B. The plant is still slated to deliver power to Microsoft data centers under the world’s first fusion power purchase agreement, signed in 2023.
• Commonwealth Fusion Systems (CFS) applied to PJM Interconnection, the largest wholesale electricity market in the country, to connect its planned ARC plant to the grid, a step the company frames as proof fusion can move beyond lab science into real deployment. The site, now named the Fall Line Fusion Power Station, has its first customer locked in: Google has agreed to purchase half of ARC’s 400 megawatt output when the plant comes online in the early 2030s.
• TAE Technologies leapfrogged an entire planned generation of its reactor. Its Norm device, which forms and sustains plasma using neutral beam injection alone, performed well enough that TAE skipped its previously planned intermediate machine and moved straight toward Copernicus, its next net energy demonstration reactor. In May 2026, TAE Beam UK, a joint venture with the UK Atomic Energy Authority, was formally established and fully funded at the UKAEA’s Culham campus to commercialize the underlying particle accelerator technology.
• Type One Energy, Tokamak Energy, and AECOM are teaming up to develop the UK's first private-sector-funded fusion power plant.

Why it matters

Fusion solves AI’s energy bottleneck. Unlimited clean energy would remove the primary constraint on computing advancement, revolutionize manufacturing, and enable breakthroughs from desalination to space exploration.

The fact that hyperscalers like Microsoft and Google are now signing offtake agreements, not just funding research, is the clearest signal yet that fusion has moved from a bet on physics to a bet on delivery timelines.

Strategic leaders should model scenarios where baseload energy costs drop 70 to 80 percent, and should watch grid interconnection filings and regulatory approvals as the leading indicators of which companies are actually closing the gap between demonstration and deployment.

That said, a bet on delivery timelines is still a bet. Fusion has a long record of slipping deadlines, the offtake deals are milestone-based, and hard problems remain in cost, fuel supply, and materials that can survive sustained operation. The optimism is warranted, but the gap between demonstration and deployment is where these timelines tend to stretch.

Fusion Breakthroughs: Commercial fusion enters the transition zone was originally published in Megatrends by HP on Medium, where people are continuing the conversation by highlighting and responding to this story.

We’re taking a look at the trends that matter most for leaders navigating the next 12–36 months.

We organize trends into three categories based on their maturity and impact timeline:

Game Changers — Trends reshaping entire industries right now. These affect how businesses operate, how value is created, and how competitive advantage is built.

Foundational Breakthroughs — Scientific and engineering advances unlocking new possibilities. These create strategic optionality for the decade ahead.

Weak Signals — Early indicators of transformations that will dominate the late 2020s. These require positioning now, even if mainstream adoption is years away.

Today, we’re diving into a Foundational Breakthrough that could redefine what AI is capable of.

Living AI: Models That Learn As They Go

Every AI model you’ve used, from the chatbot on your phone to the copilot in your enterprise software, knows only what it was taught during training. Ask it about something that happened after its cutoff date, or something outside its training data, and it simply doesn’t know. It’s frozen in time.

That’s about to change.

The next frontier in AI is about models that keep learning after deployment. Continuous, real-time learning would let AI systems absorb new information, adapt to shifting environments, and refine their own reasoning without waiting for the next training run. It’s a small conceptual shift with enormous implications: it’s widely considered one of the key unlocks on the path toward AI superintelligence.

Where This Is Emerging

Liquid AI is rethinking the neural network itself. An MIT spinout founded by Ramin Hasani and Daniela Rus, it builds “liquid” networks from neurons whose underlying equations keep changing at inference time, so the model adapts to new data on the fly instead of staying frozen after training — an architecture inspired by the 302-neuron nervous system of the C. elegans worm. The efficiency gains are striking: in one autonomous-driving test, just 19 liquid neurons matched the lane-keeping performance of a conventional network with more than 100,000 neurons, while drawing only 1 to 5 watts of power.

Safe Superintelligence (SSI) represents a new class of AI company built around a singular focus: systems that can learn and improve continuously, without compromising on safety, alignment, or human oversight. Founded in June 2024 by former OpenAI chief scientist Ilya Sutskever — co-creator of GPT-3 and GPT-4 — alongside Daniel Levy and Daniel Gross, SSI is a bet that today’s dominant recipe of more compute and more data will eventually plateau, and that reaching superintelligence will require fundamentally new ways of learning. It is a pointed wager from the person who did as much as anyone to prove that scaling works. With roughly 20 employees, no product, and no revenue, SSI has been valued at $32 billion — the market pricing a thesis rather than a business, and a signal that some of the smartest capital in AI believes continual, self-improving learning is the real path forward.

AMI Labs takes yet another route. Founded by Turing Award winner Yann LeCun after his departure from Meta, it is pursuing “world models” through an approach called JEPA — AI that learns how the physical world actually works, its cause, effect, and physics, rather than just the statistical patterns of text. The idea has drawn serious backing: a $1 billion seed round and a $4.5 billion valuation as of March 2026, from investors including Nvidia, Samsung, Jeff Bezos, and Eric Schmidt.

Why It Matters

Continual learning is the missing piece for AI superintelligence. Today’s models are static snapshots that are remarkably capable, but locked in place the moment training ends. A model that learns continuously can do something fundamentally different: discover new insights on its own, adapt to environments its creators never anticipated, and potentially develop reasoning capabilities that go beyond what it was explicitly trained to do.

That’s the shift from AI as a tool that recalls information to AI as a system that can genuinely innovate and invent.

For leaders, the strategic question is how quickly and where the first real advantages show up. Companies building products, security systems, or research tools on top of adaptive AI today are positioning themselves for a very different competitive landscape tomorrow.

Living AI: Models That Learn as They Go | 2026 Trends was originally published in Megatrends by HP on Medium, where people are continuing the conversation by highlighting and responding to this story.

We’re taking a look at the trends that matter most for leaders navigating the next 12–36 months.

We organize trends into three categories based on their maturity and impact timeline:

Game Changers — Trends reshaping entire industries right now. These affect how businesses operate, how value is created, and how competitive advantage is built.

Foundational Breakthroughs — Scientific and engineering advances unlocking new possibilities. These create strategic optionality for the decade ahead.

Weak Signals — Early indicators of transformations that will dominate the late 2020s. These require positioning now, even if mainstream adoption is years away.

Today, we’re diving into soverign stacks.

For decades, technology supply chains were optimized for efficiency. Design here, fabricate there, assemble somewhere else, ship everywhere. That model is being rewritten because efficiency is no longer the only thing governments are optimizing for.

Why now

This isn’t theoretical anymore. In just the past few months, escalating export controls, chip bans, and abrupt platform restrictions have made clear that shared global technology platforms can no longer be taken for granted, and governments are moving fast to reduce that exposure. Three pressures are converging behind the shift. Supply chain fragility, exposed during the pandemic, made clear how concentrated chip production was and how vulnerable every downstream industry was as a result. National security concerns followed close behind, as AI capabilities came to be seen less as a commercial advantage and more as a strategic asset. And economic competitiveness rounded it out: countries that can train and run their own AI models, on their own chips, are positioning to export that capability to others.

Countries are now building independent technology stacks from the ground up, from semiconductor fabs to the AI training infrastructure that runs on top of them. Since 2021, government incentives in the US have helped catalyze nearly $400 billion in announced semiconductor investments, with major production facilities breaking ground in Arizona, Ohio, and across Europe under parallel initiatives. The scale of capital being committed to domestic chip production hasn’t been seen in a generation.

The result is a layered build-out. Fabs at the bottom. Cloud and compute infrastructure in the middle. Sovereign AI models on top. Each layer reinforces the case for the others.

Recent movements

A few signals show how far this has already progressed.

China continues to invest heavily in domestic AI alternatives. DeepSeek is reportedly developing its own AI chip, aimed at inference rather than training, in a move that would reduce its dependence on Nvidia and Huawei silicon, a notable shift for a company built on squeezing more out of other people’s hardware. Alongside this, Harmony OS represents a parallel effort to reduce dependence on Western operating systems at the device level.

India is pursuing a similar strategy through BharatGen, a national generative AI initiative backed by the country’s Department of Science and Technology. The program has trained models on thousands of GPUs and introduced a 17-billion-parameter multilingual model built specifically for India’s languages and governance needs. More recently, BharatGen was named the anchor for India’s participation in Project Tapestry, a global AI Alliance consortium that lets nations fine-tune a shared base model on their own sovereign data while retaining full ownership of the result, an explicit hedge against depending indefinitely on AI systems trained on foreign contexts and governed elsewhere.

Europe and South Korea are moving at similar speed. The European Commission’s new technology sovereignty package pairs a “Chips Act 2.0” with a Cloud and AI Development Act aimed at building out domestic compute and reducing reliance on non-European providers. South Korea, meanwhile, has laid out more than $1 trillion in combined public and private investment behind what its government calls a “triple axis” strategy spanning semiconductors, physical AI, and data centers, with Samsung and SK Hynix anchoring new fabs at home.

And in the US, the shift is as much institutional as it is industrial. The agency formerly known as the US AI Safety Institute has been reformed into the Center for AI Standards and Innovation, with a mandate that includes representing US interests internationally and pushing back against what it considers burdensome foreign regulation of American AI technology. CAISI has since expanded its pre-deployment testing agreements to Google DeepMind, Microsoft, and xAI, alongside existing partners OpenAI and Anthropic, and has now completed more than 40 model evaluations, some in classified environments. The renaming reflects a broader pattern: countries aren’t just building their own technical stacks, they’re building the standards and institutions to defend them.

Why it matters

Technology, geopolitics, and economic leverage are converging into a single conversation. A chip is no longer just a chip, and a foundation model is no longer just a product. Each is also a statement about whose infrastructure, whose standards, and whose rules a country plans to operate under.

For global businesses, this means navigating an increasingly nationalized landscape, where the device manufacturer, cloud provider, the AI model creator, and even the operating system a customer uses may be shaped less by what’s best and more by where it was built. A strategy that assumes a single global technology stack is a blind spot.

Sovereign Stacks: Nations Build Independent AI and Chip Ecosystems was originally published in Megatrends by HP on Medium, where people are continuing the conversation by highlighting and responding to this story.

AI is no longer a service you call. It’s becoming a continuously operating network you inhabit.

From Monolithic Models to Coordinated Networks

The mental model most people still carry (you send a prompt, a large model somewhere in a data center responds) is already becoming obsolete. What’s replacing it is more interesting and more complex: networks of specialized models, distributed across edge devices and cloud infrastructure, routing tasks to wherever they can be handled most efficiently.

A smaller, faster model on your laptop handles the quick, privacy-sensitive work. A larger model in the cloud takes over when more significant reasoning becomes needed. Neither model operates in isolation. Together, they behave like something neither could be alone.

The infrastructure for this is being built right now, and the early results are striking. Research into model routing suggests these approaches could reduce inference costs by up to 85% while simultaneously improving accuracy. Not a tradeoff, but a genuine improvement on both dimensions at once.

Real-World Momentum

What this shift requires — and what has been missing until recently — is an orchestration layer that coordinates the network. Not just a device or a cloud service, but an intelligence that sits at the center of your data and devices and makes the whole system coherent.

HP IQ is built to be exactly that. Described by HP as a powerful AI orchestrator, it is designed as a workplace intelligence layer that coordinates experiences across HP AI PCs, workplace devices, and spaces. It processes many experiences locally on-device — including sensitive files, even offline — for speed, privacy, and lower AI costs, while routing select tasks to the cloud only when permitted by enterprise policy and user permissions. In a physical office, HP NearSense “reads the room” to create frictionless connections across select HP PCs, printers, and Poly devices, helping work move more seamlessly across nearby devices and shared spaces.

This is the architecture the hybrid AI vision requires. Not just edge and cloud living side by side, but a connected, coordinated system where intelligence flows to wherever it’s needed most.

The shift is happening across every layer of the stack.

For developers, tools like Arcee Conductor abstract this complexity entirely, enabling seamless switching across providers without changing a line of code. The underlying routing logic can evolve (new models, new cost structures, new capabilities) while the application layer stays stable.

And at the mobile edge, Google’s Gemini Nano is purpose-built to run directly on Android devices, handling on-device inference for tasks that don’t require the full weight of cloud compute, while leaning on larger Gemini models when they do. The seam between local and cloud becomes nearly invisible to the user.

Why Routing Is the New Core Competency

What unifies these developments is a shift in where intelligence lives. In the first generation of enterprise AI, the question was which model to use. The emerging question is which model to use for this specific task, right now, given these constraints.

That’s a routing problem, and solving it well turns out to be enormously valuable.

Routing well means understanding not just the capabilities of different models, but the cost curves, latency requirements, data residency constraints, and accuracy tradeoffs for every class of task your organization runs. It means building the observability to know when routing decisions are going wrong. It means treating the model layer as infrastructure — something you optimize and tune continuously.

The organizations that get this right will run AI that is faster, cheaper, more private, and more accurate than competitors using a single-model approach. The gap between a well-routed hybrid architecture and a naive one isn’t marginal. The research suggests it’s potentially an order of magnitude.

Strategic Implications for Organizations

Most enterprise AI deployments today are still built around a single model from a single provider. It’s simpler to procure, simpler to reason about, and simpler to explain to stakeholders. But it’s increasingly a ceiling.

Design for heterogeneity. The most capable AI architectures in 2026 are portfolios of models with different strengths, deployed against the tasks each handles best. The goal is optionality and composability, not consolidation around one provider.

Build for resilience. A distributed architecture isn’t just more efficient — it’s more durable. When AI processing is distributed across edge and cloud, there’s no single point of failure. If cloud connectivity drops, local models keep running. For organizations where AI is embedded in critical workflows, that’s not a secondary benefit. It’s a requirement.

Let privacy become a product feature. When sensitive data can be processed entirely on-device, without touching external infrastructure, that’s not just a compliance checkbox. It’s a genuine differentiator for customers in regulated industries, for employee trust, and increasingly for competitive positioning. HP IQ is explicit about this: processing locally improves data control and removes exposure to cloud-side risks.

Build routing as an organizational capability, not just a technical feature. Deciding which tasks go where requires a combination of technical infrastructure and operational discipline. The companies building this capability now are accumulating an advantage that will compound.

What Still Needs to Mature

The vision is compelling. The execution is harder.

Coordinating multiple models across edge and cloud introduces latency, consistency challenges, and failure modes that don’t exist in a single-model architecture. The tooling for observing, debugging, and optimizing these systems is still young.

Standards for interoperability are underdeveloped. Switching between cloud providers at the routing layer is relatively tractable. Switching between edge AI runtimes (the model formats, hardware abstractions, and deployment toolchains) is not. The lack of standardization here will slow enterprise adoption and create new forms of lock-in at the infrastructure layer.

The skill gap is significant. Building and operating hybrid AI systems requires capabilities that most engineering teams don’t yet have at scale, including model evaluation, inference optimization, routing logic design, and edge deployment. The gap between what architecture enables and what most organizations can build remains wide.

Cost transparency remains a challenge. The promise of hybrid AI is a lower total cost. But realizing that promise requires instrumentation that most organizations don’t yet have: clear visibility into what each inference costs, where it runs, and whether the routing decisions being made are actually optimal. Without that visibility, the savings are theoretical.

Looking Ahead

A genuinely new computing paradigm is taking shape underneath the product announcements and routing benchmarks.

For most of the history of software, compute was a place you sent work. The cloud was the destination. The edge was where you collected data before sending it to a more capable location.

That model is inverting. The edge is becoming a peer. The cloud is becoming a specialist you escalate to. And the intelligence that coordinates between them is itself becoming a layer of the stack worth designing carefully.

HP’s framing at Imagine 2026 put it plainly: the goal is to deploy the right compute in the right place at the right time. That’s not a product positioning statement. It’s a description of what enterprise AI infrastructure will need to do at scale.

The organizations building toward that architecture now, with routing capabilities, edge-ready hardware, and the governance to manage context and data residency across both, are building something that compounds. The ones waiting for the pattern to fully stabilize before investing are likely to find the gap has widened in the meantime.

The Networked Mind: Hybrid AI Spans Edge and Cloud was originally published in Megatrends by HP on Medium, where people are continuing the conversation by highlighting and responding to this story.

We’re taking a look at the trends that matter most for leaders navigating the next 12–36 months.

We organize trends into three categories based on their maturity and impact timeline:

Game Changers — Trends reshaping entire industries right now. These affect how businesses operate, how value is created, and how competitive advantage is built.

Foundational Breakthroughs — Scientific and engineering advances unlocking new possibilities. These create strategic optionality for the decade ahead.

Weak Signals — Early indicators of transformations that will dominate the late 2020s. These require positioning now, even if mainstream adoption is years away.

Today, we’re diving into stablecoins.

What are stablecoins?

Stablecoins are digital currencies whose value is pegged to traditional assets such as the U.S. dollar. Unlike cryptocurrencies that can experience significant price volatility, stablecoins are designed to maintain a stable value, making them increasingly useful for payments, treasury management, cross-border transactions, and global settlement.

Stablecoins have moved from a crypto-native experiment to core financial plumbing. In 2024, total stablecoin transfer volume reached $27.6 trillion, surpassing the combined transaction volume of Visa and Mastercard by more than 7%.

That number deserves a caveat. A meaningful share of stablecoin volume is trading and bot activity rather than commerce. But even accounting for that, the underlying trend is real: from corporate treasury teams moving millions across entities to gig platforms paying contractors overseas, stablecoins have become the rails businesses reach for when speed and cost matter.

Why are businesses adopting stablecoins?

The shift is being driven by three converging forces.

Settlement speed: stablecoin transactions clear in seconds, while cross-border wires can take days and pass through multiple correspondent banks along the way.

Liquidity management: businesses operating across currencies and time zones want to hold and move value without waiting for banking hours or absorbing FX spreads at every hop.

Competitive pressure is closing the loop. Once one player in a supply chain or marketplace adopts faster settlement, everyone connected to them feels the gap.

Major financial institutions are reading these signals the same way. Stablecoins offer:

• 24/7 settlement
• Lower-cost cross-border payments
• Reduced dependence on correspondent banking networks
• Greater treasury flexibility
• Programmable payment infrastructure

That combination is difficult to replicate with legacy rails built for a banking-hours world.

Recent momentum

A few data points show how quickly this is moving from theory to infrastructure:

Circle’s IPO signals institutional confidence
Circle’s IPO in June 2025 saw shares triple on opening day, valuing the USDC issuer at more than $18 billion. That kind of public market reception signals institutional investors are pricing in stablecoins as durable infrastructure, not a speculative side bet.

Payment platforms are embedding stablecoin rails
Mainstream payment players are building stablecoin rails directly into commerce. PayPal’s PYUSD brings a dollar-pegged stablecoin into one of the most widely used consumer wallets, and Stripe has introduced stablecoin payments for subscriptions, putting recurring revenue businesses on settlement rails that don’t sleep on weekends.

Embedded finance is making stablecoins invisible
The embedded finance layer is catching up fast. Highnote is enabling businesses to fold stablecoin payments directly into their own apps, which means the rails are becoming invisible. Companies won’t need to build crypto expertise in-house; they’ll inherit it through the platforms they already use.

Regulation is catching up
Just as important, regulatory frameworks around digital assets continue to mature globally. As compliance standards become clearer, enterprises gain greater confidence that stablecoin-based payment infrastructure can operate within existing financial, reporting, and risk-management requirements.

Why it matters

The pattern here echoes other infrastructure shifts: the technology becomes boring exactly when it becomes important.

Most consumers will never know whether a payment settled through ACH, SWIFT, a card network, or a stablecoin rail. What matters is that the experience becomes faster, cheaper, and more reliable.

As stablecoin infrastructure becomes embedded into payment platforms, marketplaces, ERP systems, and treasury workflows, the technology itself fades into the background.

The dollars look the same.

The receipts look the same.

What changes is what happens underneath, where settlement that once took days now takes seconds and where payment rails are increasingly programmable.

Looking forward: digital dollars become default settlement infrastructure

By late 2026, expect a growing share of global commerce to settle on-chain, often without consumers or even many businesses noticing the difference.

Looking further ahead, stablecoins are likely to become a foundational layer of global payments infrastructure, sitting alongside traditional banking systems rather than replacing them outright.

Just as most businesses today don’t think about the internet protocols that power their applications, many won’t think about the blockchain infrastructure powering their payments. They will simply expect money to move instantly, globally, and continuously.

For leaders watching this space, the question isn’t whether to have an opinion on crypto.

It’s whether your payment, treasury, and partner integration strategies assume settlement still takes days, when a growing share of the world is moving to one that doesn’t.

Stable Value Rails: Stablecoins are Rewriting the Default Global Infrastructure was originally published in Megatrends by HP on Medium, where people are continuing the conversation by highlighting and responding to this story.

For decades, the dominant promise of technology was productivity. Do the same work faster. Automate the repetitive. Reduce friction at the margins. It was a compelling proposition, and it delivered real value. But it left the fundamental nature of work largely unchanged: humans still executed, and technology assisted.

Now, that relationship is flipping.

As AI agents take on full execution of complex, multi-step workflows, human value is moving upstream. Planning, judgment, creativity, ethical reasoning, and strategic direction are no longer just differentiators. They are becoming the primary deliverables.

We are entering the Agency Economy: a moment where human agency, the ability to define what work should accomplish rather than how to accomplish it, becomes the core professional competency.

From Execution to Direction

The clearest signal of this shift is the speed at which agentic AI is moving from experimentation to operations. According to Deloitte’s 2025 survey of nearly 2,000 executives, 57% are already using agentic AI in some capacity, with deployment expanding beyond pilots into customer service, fraud detection, IT operations, and knowledge management.

The shift is significant. Teams that previously spent most of their time on task execution are now spending it on goal-setting, constraint definition, and outcome evaluation. The question has changed from “how do we complete this workflow?” to “what should this workflow accomplish, and how do we know when it’s done well?”

This represents a different cognitive mode, and most organizations haven’t designed for it.

What Changes, and What Doesn’t

In the Agency Economy, several things become more valuable: the ability to decompose a complex goal into actionable agent instructions; the judgment to recognize when an AI-generated output is technically correct but strategically wrong; the creativity to frame problems in ways that unlock novel solutions; and the accountability to own outcomes that AI helped produce.

Execution speed on routine, well-defined tasks is becoming less valuable. Those tasks matter, but AI agents increasingly handle them with accuracy and scale that outpace human throughput.

What this looks like in practice:

• A marketing team no longer writes every content variant. They define brand parameters, audience segmentation logic, and quality standards, then direct agents to generate, test, and iterate at scale.
• A procurement team doesn’t process every vendor communication. They set negotiation guardrails and relationship priorities, then direct agents to handle the execution layer.

In each case, the human contribution is upstream. The work of directing is harder than it looks, and far more consequential than the execution work it replaces.

The ROI Gap Is a Design Problem

Despite rapid adoption, the Deloitte survey found that only 10% of organizations using agentic AI currently report significant ROI. Most expect returns within one to five years. That’s a design problem.

Organizations are deploying agents into workflows built for human execution. The instructions are ambiguous. The success criteria are undefined. The escalation paths are unclear. Agents operating in that environment do what any intelligent system does when given poor direction: they optimize for the wrong thing, or they stall.

The Agency Economy only delivers on its promise when organizations redesign roles and workflows around the assumption that humans are directors. That means being explicit about what good outcomes look like, building feedback loops that let agents learn from human judgment, and creating governance structures that keep humans accountable for what agents produce.

This is the organizational challenge that separates the organizations capturing value from those still waiting for it.

Strategic Implications

1. Redefine roles around judgment, not task completion. Job descriptions built around execution will need to be rebuilt around direction. The question for every function is: what decisions here genuinely require human judgment, and what can be specified well enough to delegate to an agent? Organizations that answer this question deliberately will move faster than those that let it happen to them.
2. Invest in the human skills that agents can’t replace. Critical thinking, ethical reasoning, cross-functional communication, and the ability to evaluate AI outputs against strategic intent are not soft skills. They are the core competencies of the Agency Economy. L&D programs that focus on AI tool adoption without developing these upstream skills are solving the wrong problem.
3. Design agentic workflows for auditability, not just automation. When agents are executing on behalf of humans, the humans remain responsible for outcomes. That requires clear records of what instructions were given, what outputs were produced, and where human judgment overrode agent recommendations. Organizations that build this visibility in from the start will have a significant governance advantage as regulatory frameworks mature.
4. Build the hardware layer that makes agency possible. Agentic AI workflows generate a different kind of compute demand than prompt-response interactions: they require sustained, multi-step reasoning, often on sensitive organizational data. Running those AI workflows locally at the edge, rather than routing everything to the cloud, addresses both the performance and the data sovereignty requirements that enterprise adoption demands.

HP’s latest AI PCs are built with NPUs designed to support agentic AI workloads locally, enabling fast inference without cloud dependency. HP IQ runs a 20-billion-parameter model on-device, handling routine task execution so knowledge workers can stay focused on the direction layer.

Recently, our team was tracking edge inference trends, including how AI workloads are shifting from cloud to on-device processing. We’d been capturing and tagging relevant articles, podcasts, and analyst reports on this trend.

Our Intel Digest agent, which synthesizes research into actionable intelligence, picked up those highlights, scored them against HP’s strategic priorities, and clustered them with related signals it had already been tracking. It revealed that several startups already in our pipeline were part of the same emerging opportunity, connecting companies we hadn’t previously linked. It generated a themed briefing, created a standalone Insight note linking the trend and related startups to specific HP business units, and produced follow-up actions: update the relevant startup profiles, flag startups for our AI PC team, and monitor others for investment outreach.

There is a workforce dimension here, too. HP’s 2025 Work Relationship Index found that only 20% of knowledge workers report a healthy relationship with work. A properly-designed Agency Economy is part of the answer. When people spend less time on execution and more time on work that genuinely requires their judgment, the nature of that relationship changes.

What Needs to Mature

The Agency Economy is a compelling direction, but several things need to develop in parallel for it to reach its potential.

Role redesign at scale. Most organizations are still adding agents to existing workflows rather than redesigning workflows around agent capabilities. The productivity gains remain incremental until the redesign happens. That requires organizational will, not just technical investment.

Agent governance frameworks. As agents take on more consequential work, the question of who is responsible for what they produce becomes urgent. Clear accountability structures, audit trails, and escalation protocols are not optional features. They are prerequisites for operating at scale especially in regulated environments.

The direction skills gap. Instructing an AI agent well is a skill. Knowing when to override it is a skill. Evaluating whether an output is technically correct but strategically wrong is a skill. Most organizations don’t yet explicitly train for these capabilities, and the gap is showing up in adoption results.

Compute accessibility. The Agency Economy’s benefits currently accrue most strongly to organizations with the budget and infrastructure to deploy agentic workflows at scale. As on-device AI matures and the cost curve compresses, that access will broaden, but the pace matters. Organizations that build the infrastructure foundation now will have a meaningful head start.

Looking Ahead

The reframe is: it’s not about humans doing less. It’s about humans doing fundamentally different things.

The work of direction, setting goals, defining constraints, evaluating outcomes, making judgment calls that require context and values and accountability, has always been what organizations pay most for.

This is the moment when the infrastructure catches up to that reality. The execution layer becomes reliable enough to delegate. The direction layer becomes consequential enough to deserve full human attention.

The organizations that will lead this transition are the ones designing for it now: rebuilding roles around judgment, investing in the human skills that agents can’t replicate, and building the compute infrastructure that makes agentic workflows fast, private, and controllable.

The shift from doing to directing is already underway. The question is whether you’re redesigning around it or waiting to react.

The Agency Economy: Why Work Is Shifting From Doing to Directing was originally published in Megatrends by HP on Medium, where people are continuing the conversation by highlighting and responding to this story.

If there’s one shift that defines AI’s evolution in 2026, it’s the move from amnesia to understanding. We’re leaving behind assistants that forget everything between sessions and entering an era where AI genuinely learns who you are, how you work, and what you’re trying to accomplish.

From Goldfish Memory to Institutional Knowledge

The frustration with early AI tools was real: you’d have a useful conversation, close the window, and come back the next day to explain everything from scratch. It was like working with the world’s most knowledgeable colleague who had no short-term memory.

Today’s AI systems are developing what researchers call “persistent understanding,” which is the ability to maintain continuity across conversations, remember preferences, context, and accumulate knowledge about your work, your team, and your goals over time.

The shift is driven by three technical advances working in concert: massive context windows that let AI reason across entire document libraries or codebases in a single session; persistent memory architectures that retain and retrieve information across interactions; and improved retrieval systems that surface the right context at the right moment.

How This Changes How We Work

The clearest near-term impact is the elimination of repetitive setup work. Instead of re-explaining your priorities, preferences, or project history at the start of every interaction, AI systems learn and retain that context and act on it. Organizations must redesign workflows around AI that already knows your context, not AI that waits for commands.

• Productivity compounds: No need to reexplain goals, tone, or priorities. Knowledge builds session over session, so rather than retraining AI you can jump in and quickly perform tasks.
  
  This is the core premise behind Mem0: AI agents forget, but purpose-built memory infrastructure remembers, enabling personalized experiences that get sharper over time.
• Meeting continuity: AI that connects today’s discussion to decisions made three weeks ago, flagging recurring issues, understanding situational urgency, noting which action items keep coming up unresolved, and surfacing relevant history without you asking.
• Proactive briefings: Rather than searching for what you need before a board meeting or weekly review, AI that has learned your patterns assembles relevant materials in advance, the way a well-prepared chief of staff would.
  
  OpenAI Frontier is built on this premise: as AI coworkers operate, they build memories, turning past interactions into useful context that improves performance over time.
• Intelligent workflow management: Email triage, scheduling, and task coordination that adapt to your learned priorities, making judgment calls rather than following rigid rules. The difference between a tool and a collaborator.
  
  Claude Cowork takes this further, allowing you to describe an outcome, step away, and come back to finished work Claude, including formatted documents, organized files, and synthesized research, without supervising each step. The difference between a tool and a collaborator.
• Institutional continuity: When team members leave or projects shift, AI that has built up organizational context becomes a living record of how and why decisions were made, reducing the cost of transitions and knowledge loss.
• Better decision making: AI has situational awareness, understanding urgency, time of day, role, and environment. Outputs are adjusted based on audience and topics, be it executive-level or technical. Real-time constraints are taken into consideration. All so data becomes immediately useful not just informational.
  
  Ray-Ban Meta’s AI glasses extend this further, giving AI a continuous point-of-view perspective throughout your day, with the ability to remember where you parked, translate speech in real time, and surface contextual answers hands-free.
• Trusted advisor: As AI learns your communication style, decision preferences, and priorities it adapts to you. Anticipating your needs before explicit requests, adjusting outputs for tone, style, layout and situation.
  
  Zuckerberg describes this as personal superintelligence: AI that can see what you see, hear what you hear, and talk with you throughout the day, improving your memory and decision-making without pulling you out of the moment.

The common thread: these systems aren’t executing scripts. They’re making small but meaningful judgment calls based on accumulated context, better understanding, and improving with every interaction.

Strategic Implications for Organizations

Most companies are still deploying AI as if they’re adding calculators to their toolkit. Context-aware AI requires a different mental model: persistent infrastructure that accumulates organizational knowledge over time.

1. Design for continuity, not just capability.

Workflows should assume AI already knows relevant context. If you’re still building processes where users need to re-explain background information each time, you’re underutilizing what these systems can do. The design question shifts from “what can AI do?” to “what does AI already know, and how do we build from there?”

2. Treat contextual data as a strategic asset.

An AI system that understands your organization’s judgment calls, preferences, and institutional quirks becomes increasingly difficult to replicate and increasingly difficult to replace. Building this context intentionally, and governing it carefully, is a competitive move, not just an IT decision.

3. Get ahead of the governance challenge.

The systems that provide the most value are the ones that see the most, which creates a genuine tension between utility and privacy. Organizations that wait for regulation to force their hand will be reactive. The winners will establish clear frameworks now: what context AI should and shouldn’t retain, how long it holds information, and how individuals maintain meaningful control.

4. Adopt Edge AI for context ownership and cost efficiency

Move inference closer to data sources to maintain context ownership and data sovereignty, enhance privacy, lower latency, and reduce compute costs, creating a secure architecture with better performance and scalability. This approach is increasingly reflected in industry solutions such as HP’s Edge AI strategy, which integrates NPUs into PCs and workstations to enable secure, low-latency local inference on devices like AI-enabled notebooks and edge AI inference systems.

One dynamic worth watching: AI systems that establish themselves as the context layer early will be hard to dislodge. Switching costs rise as systems accumulate more knowledge about users and workflows, which is both a reason to choose carefully and a reason to move quickly once you’ve chosen.

Key factors that still need to mature

For this shift to reach its full potential, several things need to advance in parallel, and the progress on each will determine not just how quickly this technology spreads, but who benefits and on what terms.

• Privacy and governance frameworks: The systems that deliver the most value are the ones that see the most, creating an inherent tension that hasn’t been resolved. An AI that knows your priorities, your communication patterns, and your organization’s decision history is extraordinarily useful. It’s also an extraordinarily sensitive data store. Right now, most organizations don’t have clear policies on what AI should be allowed to retain, for how long, or under what conditions that data can be used to train future models. Regulatory frameworks are moving, but unevenly.
  
  The EU AI Act sets some guardrails, but enterprise AI memory sits in a gray zone that neither privacy law nor existing data governance policies were written to address. Until organizations can articulate clear answers to basic questions about who owns the memory layer, what happens to it when an employee leaves, and how it’s audited, adoption will remain cautious at the enterprise level, precisely where the stakes are highest.
• User trust and transparency: Even users who are intellectually comfortable with AI memory often don’t know, in practice, what their AI actually knows about them. That opacity is a real barrier. There’s a meaningful difference between an AI that has context and an AI whose context you can see, interrogate, and correct. Most current implementations offer little of the latter.
  
  Building trust here is a design philosophy question. The systems that gain the deepest adoption will likely be the ones that make the memory layer legible: showing users what’s been retained, why it was surfaced in a given moment, and how to edit or remove it. Think of it as the difference between a black-box recommendation engine and one that says, “we suggested this because you’ve done X three times this quarter.” The second builds confidence. The first eventually erodes it.
• Cross-system portability: If an AI has learned how you work, can you take that knowledge with you when you change tools or vendors? Right now, the answer is almost always no, and that’s not an accident. The organizations building these memory layers have every incentive to make them sticky. But this creates a new and particularly durable form of lock-in. Previous generations of vendor lock-in were about data: migrating your CRM, porting your email archive. Context lock-in is different.
  
  This is about whether a new system can understand the institutional logic that your previous one learned over the years. Interoperability standards for AI memory don’t yet exist in any meaningful form, and the window to establish open standards is narrowing as incumbents entrench. This is worth watching closely, both as a procurement risk and as an issue likely to attract regulatory attention.
• Compute and cost efficiency: Maintaining rich, reasoning-ready context at scale is not cheap. Every time an AI needs to retrieve, synthesize, and persist across a complex workflow, it consumes significantly more compute than a simple prompt-response exchange. Today, that cost profile limits who can fully participate. Large enterprises with substantial AI budgets can absorb it.
  
  Smaller organizations, individual knowledge workers on standard subscriptions, and use cases with thin margins largely cannot. The cost curve will compress, as it always does, but the pace matters. If the productivity gains from persistent AI context accrue primarily to well-resourced organizations for the next several years, it will widen rather than close existing capability gaps. The infrastructure buildout happening now will eventually commoditize this layer, but “eventually” is doing a lot of work in that sentence.

Looking Ahead

What’s easy to miss in all of this, underneath the technical architecture and the governance frameworks, is that something genuinely new is happening in how people relate to the tools they work with.

For most of the history of software, tools were stateless. They didn’t learn. They didn’t accumulate anything about you. Every session started from zero.

That’s changing. And while the challenges around privacy, portability, and trust are real and worth taking seriously, so is the possibility on the other side: AI that actually knows your context, that improves the more you work with it, and that handles the accumulated overhead of organizational knowledge so you don’t have to carry it alone.

We’re not there yet, fully. But the direction is clear. The assistants that forgot everything are becoming the ones that remember what matters.

The Context Shift: AI Moves from Prompts to Persistent Understanding was originally published in Megatrends by HP on Medium, where people are continuing the conversation by highlighting and responding to this story.

Corporate venture capital (CVC) has moved from the edges of innovation strategy to its center. What was once an exploratory function is now a core mechanism for how large organizations sense change, test new capabilities, and shape the markets they compete in.

The shift isn’t about doing more deals. It’s about making more intentional investments — and using them to accelerate learning, capability building, and long-term advantage. Recent venture data shows that this pattern is holding steady, even as capital flows fluctuate.

Investor trends: Higher conviction

CVC participation in venture markets has stabilized, but the way corporates deploy capital has changed meaningfully.

CVC-backed deal value reached a decade high while deal count remained consistent with recent years, signaling a shift toward fewer, higher-conviction investments. Source: PitchBook–NVCA Venture Monitor, Q4 2025, Investor Trends

Key insights:

• Deal count with CVC participation in the most recent year remains in line with the prior two years
• Deal value surged to its highest level in the past decade
• Corporations are concentrating capital with clearer strategic alignment

This pattern reflects a broader recalibration. In earlier cycles, CVC programs were often evaluated by activity volume. Today, effectiveness is measured by strategic impact — how quickly an investment translates into insight, partnership, or operational advantage.

Why CVC influence outweighs participation

While only about one-fifth of VC deals include a corporate investor, CVCs account for a significantly larger share of total deal value. Source: PitchBook–NVCA Venture Monitor, Q4 2025, Investor Trends

Key insights:

• Roughly one in five VC deals includes a CVC
• CVCs account for a larger share of total capital deployed
• Strategic investors are anchoring larger rounds, especially in platform and infrastructure plays

The implication is subtle but important: CVC is no longer following venture markets. In many cases, it is shaping them — by directing capital toward technologies that align with enterprise-scale adoption and long-term transformation.

Industry-Specific CVC models and how they differ[AB1]nbsp;

CVC is not a one-size-fits-all strategy. How it works — and what success looks like — varies significantly by industry.

Healthcare and Life Sciences CVC: Long horizons, deep integration

Healthcare CVC models are built around long development timelines, regulatory complexity, and platform thinking. Investments often focus on technologies that can support multiple therapeutic areas or operational workflows over time.

How the model differs:

• Longer investment horizons
• Heavy emphasis on clinical validation and regulatory pathways
• Strategic value measured in integration, not speed

Industry examples:

• Johnson & Johnson Innovation has used minority investments to build early access to surgical robotics, digital surgery platforms, and AI-enabled clinical tools — many of which later evolved into acquisitions or deep partnerships.
• Roche Venture Fund has consistently backed data infrastructure and diagnostics platforms that complement Roche’s core R&D engine, accelerating time from discovery to deployment.

In healthcare, CVC acts as an extension of R&D, not a replacement for it.

Hi-Tech CVC: Platform Extension and Ecosystem Intelligence

In the technology sector, CVC operates at the intersection of product strategy and market intelligence. Investments are rarely about financial return alone. They are about extending platform reach, accessing emerging capabilities before they reach maturity, and building ecosystem influence in markets that move faster than traditional R&D cycles allow.

The best hi-tech CVC programs function as a real-time signal layer for the enterprise — surfacing where technology is heading before the rest of the market can see it clearly.

How the model differs:

• Shorter investment cycles with faster feedback into product and strategy
• Strong emphasis on AI infrastructure, developer platforms, and ecosystem plays
• Strategic value measured by capability access and platform extensibility
• Deep integration with internal business units to accelerate adoption of emerging tools

Industry examples:

• HP Tech Ventures has built our model around early-stage investments in startups at the frontier of AI, the future of work, and human-centric computing. Rather than chasing broad portfolio volume, HP Tech Ventures concentrates on companies whose technology can extend HP’s core platforms or unlock entirely new capabilities across its product and services ecosystem. The program pairs minority investment with direct collaboration, creating a structured path from external innovation to internal impact — and, increasingly, using AI agents internally to accelerate startup sourcing, market analysis, and portfolio intelligence at scale.
• Intel Capital has invested across semiconductor software, cloud infrastructure, and AI acceleration — consistently backing companies that expand the addressable market for Intel’s core platform and deepen its developer ecosystem.

In hi-tech, CVC is not a passive bet on the future. It is an active mechanism for staying close to the frontier while competitors are still reading about it.

Financial Services and Fintech CVC: modernizing the core

In financial services, CVC is tightly linked to internal transformation. Rather than chasing disruption, banks and insurers invest in startups that modernize core systems and enable new operating models.

How the model differs:

• Shorter feedback loops
• Success tied to internal adoption
• Strong alignment with risk, compliance, and data strategy

Industry examples:

• Citi Ventures operates as Citigroup’s corporate venture arm, making minority investments in fintech, data, and infrastructure companies while working directly with internal business units to pilot and scale new capabilities.
• BNY Mellon SPIN program uses corporate venture capital to modernize custody, asset servicing, and data infrastructure, pairing minority investments with direct collaboration across its operating businesses.

Here, CVC functions as a modernization accelerator, not an external skunkworks.

Manufacturing and Industrial CVC: From pilot to production

Industrial CVC models prioritize applicability. Investments focus on technologies that can be tested quickly and scaled across operations — automation, robotics, AI-driven quality control, and advanced materials.

How the model differs:

• Strong emphasis on pilots and proof-of-value
• Deep collaboration with engineering and operations teams
• Value measured in efficiency, resilience, and throughput

Industry examples:

• Bosch Ventures consistently backs industrial AI and sensor technologies, often integrating them into Bosch’s own manufacturing footprint.

In this sector, CVC bridges the gap between experimentation and execution.

Industry-specific CVC success patterns

Across industries, successful CVC programs share common characteristics:

• Clear strategic mandate: Investments map directly to future capabilities the enterprise needs
• Defined pathways to impact: Pilots, partnerships, or acquisition routes are planned early
• Operational ownership: Business units are engaged, not surprised

CVC works best when it is embedded in how decisions get made — not when it operates as a standalone investment arm.

2026 CVC Predictions: How corporate venture capital is evolving

Looking ahead, several patterns are becoming clearer across industries.

💡 Prediction 1: CVC Portfolios Will Continue to Narrow

Corporates will make fewer investments — but commit more deeply. Portfolio breadth is giving way to conviction and focus.

💡 Prediction 2: Infrastructure Will Outpace Applications

CVC dollars are increasingly flowing toward AI infrastructure, data platforms, automation layers, and integration tools — the technologies that enable multiple downstream use cases.

💡 Prediction 3: CVC Will Formalize as an M&A Pathway

Minority investments will increasingly serve as structured on-ramps to acquisition, reducing integration risk and accelerating strategic alignment.

💡 Prediction 4: Strategic Impact Will Replace Financial Optics

Success will be measured less by paper returns and more by:

• Speed to deployment
• Capability gained
• Organizational learning

CVC is becoming a core learning system for enterprises navigating uncertainty.

CVC as a strategic operating capability

Corporate venture capital has matured into a disciplined, strategic function. The data is clear: fewer deals, larger commitments, and sharper intent. While execution differs by industry, the direction is consistent.

The organizations that succeed won’t be the ones that invest the most. They’ll be the ones that connect venture investing directly to how work gets done, how platforms evolve, and how new capabilities take root.

In that sense, CVC is no longer about betting on the future — it’s about building it.

Why Corporate Venture Capital Is Becoming an Industry-Defining Strategy was originally published in Megatrends by HP on Medium, where people are continuing the conversation by highlighting and responding to this story.