Artificial intelligence has spent the past few years in a relentless race for more computing power. Bigger GPUs. Bigger clusters. Bigger data centers. Bigger electricity bills.

But the next major AI performance breakthrough may come from something less glamorous: helping processors get data from memory faster.

San Francisco semiconductor startup Volantis has raised $88 million in Series A funding to tackle exactly that problem. Rather than attempting to build yet another GPU to compete directly with NVIDIA or AMD, Volantis is developing photonic technology that uses light to connect AI computing chips with much larger pools of memory.

According to [Reuters] the company believes its optical approach could allow a GPU to communicate with as many as 220 memory chips, dramatically expanding the amount of memory that can sit within high-speed reach of an AI processor.

That could matter enormously as AI moves from model training toward always-on inference, coding agents, long-context models and autonomous enterprise workflows.

In other words, AI’s next speed limit may not be how quickly a GPU can calculate. It may be how quickly the GPU can get the data it needs.

The $88 Million Bet on the AI “Memory Wall”

Volantis says its Series A was led by Lachy Groom and Abstract Ventures, with participation from John Doerr, VXI Capital, Triatomic and Susa Ventures, alongside angel investors including Dwarkesh Patel, Naveen Rao and Sholto Douglas. The round takes the company’s total funding to approximately $97 million.

The company explains the strategy in its own post [Volantis] Rather than treating memory merely as storage sitting beside a processor, Volantis wants to rethink the physical connection between memory and AI compute.

That distinction is important.

Modern AI accelerators can perform an astonishing number of mathematical operations every second. But large language models contain enormous volumes of parameters and intermediate data that constantly need to move between memory and the processor.

When the processor has to wait for that data, expensive compute capacity can sit underutilized.

Welcome to the memory wall.

Why AI Chips Need So Much Memory Bandwidth

Think of a GPU as an extremely fast chef.

The chef might be capable of preparing hundreds of meals every hour, but that speed is useless if ingredients arrive through a tiny kitchen window one tomato at a time.

AI processors face a similar problem.

Model weights, key-value caches, context and other data have to move continually between memory and computation. As models become larger and context windows expand, the amount of information traveling through the system grows substantially.

That’s why high-bandwidth memory, or HBM, has become critical to modern AI accelerators.

Micron offers an excellent technical explanation in its [blog on how memory os becoming the backbone of the AI era] Its broader argument is worth paying attention to: AI system performance is increasingly determined not only by computation but by bandwidth, latency and data movement.

The established GPU industry is already responding aggressively.

For example, NVIDIA’s next-generation Vera Rubin platform incorporates HBM4 and is designed for up to 288 GB of memory and as much as 22 TB/s of aggregate memory bandwidth per GPU, according to the company’s technical materials. NVIDIA’s illustrates how central memory bandwidth has become to accelerator design.

Volantis is proposing a different way to push those limits.

Instead of Copper, Volantis Wants to Move AI Data With Light

Traditional chip packages predominantly move information through electrical connections.

The catch is that extremely high-speed electrical connections become increasingly difficult as distance, bandwidth, energy consumption and signal integrity requirements rise.

Volantis wants to replace some of those short electrical connections with optical links.

Its technology uses tiny lasers known as vertical-cavity surface-emitting lasers, or VCSELs, to transmit information through light rather than conventional electrical wiring.

VCSELs are not entirely experimental technology. They have already been manufactured at massive scale for applications including smartphone sensing systems. Reuters notes that related VCSEL technology has been used in Apple’s Face ID ecosystem, providing Volantis with a comparatively mature manufacturing foundation rather than requiring an entirely new laser supply chain.

The clever bit is where Volantis wants to put the technology: inside the AI accelerator’s memory architecture.

Its technical materials say electrical interposer connections typically reach only a few millimeters, while Volantis’s optical waveguides are designed to extend beyond 200 millimeters. The company says this additional reach could make it possible to connect more than 220 memory chiplets into one uniform-latency memory pool.

That is a radically different geometry for AI hardware.

Instead of cramming a small number of extremely expensive memory stacks as close as physically possible to a GPU, photonics could potentially create a much larger neighborhood of memory surrounding the compute engine.

What Makes Volantis Different?

Volantis is not simply talking about putting optical cables between racks of GPUs.

Its pitch goes much closer to the processor.

The company says its optical fabric uses integrated micro-VCSEL arrays and extremely small optical waveguides rather than conventional fiber and external lasers. It lists target characteristics including 24 Gb/s per lane, sub-five-nanosecond latency, tens of thousands of lanes and more than 200 TB/s of aggregate bandwidth.

Volantis also says its links consume less than one picojoule per bit and have demonstrated wafer-scale bit-error rates below 1e-12. These figures come from Volantis itself and should therefore be understood as company-reported technical results and design characteristics rather than independent third-party benchmarks.

That distinction matters because semiconductor engineering has a habit of turning beautiful diagrams into very difficult manufacturing problems.

Packaging yields, thermal behavior, laser reliability, software compatibility, manufacturing cost and large-scale system integration will all matter if Volantis wants its technology to move from promising silicon into production data centers.

The A-1 System Targets Massive AI Models

Volantis calls its first planned system A-1.

The company’s announcement says A-1 is being designed for extremely large AI inference workloads, with ambitions extending into multi-trillion-parameter models and extremely high per-user token throughput. The company’s website currently promotes support for models above 10 trillion parameters and speeds of up to 10,000 tokens per second per user, while its formal funding announcement distributed through PR Newswire describes A-1 as being designed for models exceeding 20 trillion parameters.

Those are targets, not results that prospective customers should assume have already been demonstrated in a commercial production system.

That caveat aside, the goal reveals where Volantis sees the AI market heading.

The company is focusing heavily on inference rather than simply training.

That is important because the AI economy is increasingly about running models repeatedly for millions of users and software agents—not merely training a frontier model every few months.

Why Faster Memory Could Supercharge AI Agents

Imagine an AI coding agent asked to analyze a large software repository.

The agent may need to ingest hundreds of files, reason across multiple dependencies, generate code, run tests, inspect results and revise its solution.

That workflow isn’t one simple model response. It is a long chain of inference operations.

When every step waits on model execution, latency stacks up.

Volantis argues that dramatically increasing memory bandwidth and capacity could make these workflows far more interactive. Its own funding announcement uses coding agents as an example of applications that could potentially move from lengthy workflows toward near-real-time operation if model inference becomes dramatically faster.

Energy Efficiency Could Become Just as Important as Speed

There is also a sustainability angle.

Modern AI systems spend considerable energy not just performing mathematical operations but moving information among processors, memory and networking hardware.

As infrastructure expands, optimizing data movement could become a meaningful component of reducing the energy required per AI query.

Micron’s research has similarly emphasized memory as an important frontier for AI energy efficiency, noting that memory, storage and the interconnects surrounding them can represent a substantial portion of system power depending on the workload and architecture.

Optical connections are therefore interesting for reasons beyond benchmark bragging rights.

If photonics can move substantially more information using less energy per bit, improved interconnects could help data centers generate more useful AI work from constrained electrical power.

That matters because electricity availability is rapidly becoming an infrastructure constraint of its own.

The Challenges Volantis Still Has to Prove

An $88 million funding round is a major vote of confidence. It is not the same thing as volume production.

Volantis now has to demonstrate that its architecture works reliably and economically at scale.

That means proving several things simultaneously: the photonic components must achieve high manufacturing yields; packaging must tolerate the thermal environment surrounding powerful AI processors; the memory system must deliver reliable latency and bandwidth under real workloads; software needs to take advantage of the architecture; and the final system needs to produce sufficiently compelling total-cost-of-ownership improvements to justify infrastructure changes.

The competitive landscape won’t stand still, either.

NVIDIA, AMD, memory manufacturers and advanced-packaging suppliers continue improving HBM capacity, bandwidth, chiplet architectures and scale-up networking. NVIDIA’s Rubin architecture, for example, shows just how aggressively conventional HBM-based systems are evolving.

So Volantis does not merely have to prove that photonics works.

It has to prove that photonics improves quickly enough to outrun improvements in the technologies it hopes to disrupt.

What the Volantis Funding Means for the Future of AI

The most interesting thing about Volantis’s $88 million raise is not necessarily Volantis itself.

It is what the investment says about where the AI industry believes the next bottlenecks are emerging.

The first phase of the generative AI boom revolved around models.

The second became a race for GPUs.

The next phase is becoming a full-stack infrastructure battle involving memory, networking, advanced packaging, optical interconnects, cooling, energy and highly specialized silicon.

That changes how technology leaders should think about AI strategy.

The fastest model on paper means very little if infrastructure cannot feed it enough data, power it economically or operate it at acceptable latency.

For CIOs and enterprise technology leaders, this means AI infrastructure decisions should increasingly consider memory architecture, inference efficiency, networking and cost per token, not simply which accelerator produces the largest benchmark number.

For semiconductor companies, it means some of the industry’s biggest opportunities may exist between the processor and memory rather than inside the processor itself.

For the wider AI ecosystem, Volantis provides another signal that the next generation of innovation will involve as much physics as software.

Conclusion

Volantis’s $88 million funding round highlights a major shift in the AI hardware race: the industry is no longer focused only on building faster processors. The ability to move enormous amounts of data between AI accelerators and memory is becoming just as important as raw compute performance.

By using photonic interconnects to link AI chips with larger pools of memory, Volantis is taking aim at the growing “memory wall” that can limit the performance and efficiency of advanced AI systems. If the company can successfully deliver its promised gains in bandwidth, latency, energy efficiency, and memory capacity at commercial scale, its technology could help reshape how future AI infrastructure is designed.

There are still significant technical and manufacturing challenges ahead. Volantis will need to prove that its architecture can perform reliably in real-world data centers, compete economically with rapidly improving HBM-based systems, and integrate smoothly with the broader semiconductor ecosystem.

Even so, the company’s funding sends a clear signal about where AI infrastructure innovation is heading. The next breakthroughs may come not only from more powerful GPUs, but from smarter ways of connecting compute, memory, networking, and storage.

As AI models become larger and inference workloads become more demanding, technologies that reduce data-movement bottlenecks could become critical to lowering costs, improving performance, and making advanced AI systems more accessible.

Volantis is still early in that journey, but its photonics-first approach places it squarely in one of the most important emerging battles in AI hardware: making sure tomorrow’s processors can access data as quickly as they can process it.

The artificial intelligence race is no longer just about building a faster GPU or training a larger language model. Increasingly, it is about connecting models, software, hardware, simulation, robotics, and the physical world into one technology stack.

AMD just placed an $8.2 billion bet on that future.

On September 28, 2026, Advanced Micro Devices announced that it had entered into a definitive agreement to acquire World Labs, the spatial-intelligence company led by renowned computer-vision researcher Dr. Fei-Fei Li. The transaction is an all-stock deal valued at approximately $8.2 billion and is expected to close by the end of 2026, subject to regulatory approvals and customary closing conditions, according to [AMD]

That qualification matters: AMD has agreed to acquire World Labs; the transaction has not yet completed.

Once the deal closes, Fei-Fei Li is expected to become AMD’s Executive Vice President and Chief Scientist, reporting directly to AMD Chair and CEO Dr. Lisa Su. World Labs’ researchers will continue working on advanced AI models within AMD, creating a much tighter connection between frontier AI research and the processors, accelerators, software, and computing systems on which those models run.

And that is where this acquisition becomes much more interesting than its eye-catching price tag.

What Exactly Is AMD Buying for $8.2 Billion?

World Labs is not another company building a ChatGPT competitor.

Founded in 2024 and led by Fei-Fei Li, the company is focused on spatial intelligence: giving AI systems a richer understanding of three-dimensional spaces, objects, relationships, motion, and the physical rules governing environments.

In its own announcement [World Labs] saying that pushing its research forward requires closer integration between model research, systems, and computing infrastructure. The companies had already begun working together on model training and inference optimization using AMD GPUs, meaning this is less a cold acquisition and more the deepening of an existing technical relationship.

World Labs has been developing models capable of generating, reconstructing, and simulating interactive 3D environments from text, images, and video. Its first product, Marble, can create persistent three-dimensional worlds from those inputs.

Think of the distinction this way: a language model can describe a room. A spatially intelligent model aims to understand how that room is constructed, where the objects are, how they relate to one another, what happens when something moves, and potentially how an autonomous machine should behave within it.

That capability could become important across robotics, industrial simulation, architecture, entertainment, autonomous systems, digital twins, scientific research, and many other applications involving the physical world.

Why AMD Wants World Labs

AMD has spent years strengthening its position as an alternative AI-computing platform to Nvidia. But competing in AI increasingly requires much more than manufacturing powerful accelerators.

The modern AI stack stretches from models and frameworks all the way through compilers, networking, memory, accelerators, CPUs, racks, and data centers.

Acquiring World Labs could give AMD something particularly valuable: direct access to researchers building the kinds of advanced workloads its future hardware will need to run.

Lisa Su summarized that logic in AMD’s announcement: understanding how models are evolving can help the company determine how future computing platforms should be designed. [Advanced Micro Devices, Inc.]

In other words, instead of waiting for the next generation of AI models to arrive and then optimizing hardware around them, AMD could now participate much earlier in the process.

That creates a feedback loop:

Researchers build increasingly demanding AI models → AMD sees where the computing bottlenecks are → AMD optimizes hardware and software around those requirements → researchers get better infrastructure for the next generation of models.

The Deal Signals That “World Models” Are Becoming Serious Business

A few years ago, much of mainstream generative AI revolved around language.

The next battleground may be models capable of understanding worlds.

[TechCrunch] describes world models as a broad emerging class of technologies designed to achieve a stronger understanding of physical reality. The term can encompass systems that reason about visual information as well as models capable of creating persistent, high-fidelity simulations.

That distinction could become enormously important.

Large language models learned patterns in text. Multimodal models expanded that understanding to images, audio, and video. World models take another step by attempting to represent environments, geometry, movement, interactions, and physical consequences.

A robot, for example, needs more than an ability to recognize a cup. It may need to determine where the cup is, whether it can reach it, which direction its handle faces, how much force is safe to use, which obstacles stand between the robot and the cup, and what will happen after it picks the cup up.

That is spatial intelligence.

And World Labs wants to make it scalable.

Why Robotics Could Be One of the Biggest Winners

The AMD–World Labs deal arrives at an especially interesting moment for robotics.

Modern robots are rapidly gaining better vision, reasoning, simulation, and machine-learning capabilities. Yet training robots in the real world remains slow and expensive.

If every robot must learn every failure by physically experiencing it, scaling development becomes painfully inefficient.

World models offer another path.

Researchers can create simulated environments where robots practice tasks, encounter changing conditions, make mistakes, and develop policies before deployment into real environments.

World Labs itself has described this as a real-to-sim-to-real approach. Its research explores how spatial models can reproduce real environments in simulation, use those environments for robot training and evaluation, and then transfer what the machines learn back into the physical world.

That matters because real environments are wonderfully inconvenient. Lighting changes. Objects move. Floors get messy. People walk through workspaces. Cables shift. Machines break.

AMD Is Moving Beyond the “Chip Company” Label

Perhaps the most important implication of the World Labs acquisition is what it says about AMD itself.

The company increasingly wants to compete at the system level, not simply sell individual processors.

AMD already spans CPUs, GPUs, networking, embedded computing, AI PCs, server infrastructure, and its ROCm software ecosystem. Adding a major frontier AI research organization brings models and model researchers directly into that mix.

[The Verge] highlights this broader strategy, noting that AMD intends to bring World Labs’ researchers and model expertise closer to its hardware, software, and systems development.

The potential advantage is not simply owning a world model.

It is learning how frontier models behave while AMD is designing the computers that will run them.

AI workloads are changing incredibly quickly. An architecture optimized for today’s language-model inference may face very different requirements from tomorrow’s interactive 3D simulation, robot learning, long-horizon reasoning, or real-time multimodal processing.

World Labs could give AMD an unusually close view of those emerging requirements.

The Open AI Ecosystem Could Be an Important Part of AMD’s Strategy

Both AMD and World Labs have emphasized their intention to support an open AI ecosystem.

World Labs said the combined organization plans to build an end-to-end AI ecosystem spanning hardware, software, platforms, and broadly accessible open models. AMD similarly framed the acquisition around strengthening openness across AI infrastructure.

That strategy could be important commercially.

AI developers increasingly worry about becoming locked into a single vendor’s hardware, software, model ecosystem, or cloud environment. Interoperability and open tooling can give enterprises more flexibility over where workloads run and how infrastructure is purchased.

AMD can potentially use openness as a differentiator while simultaneously building deeper integration between its own hardware and frontier AI research.

The balancing act will be making those systems genuinely open while still creating enough technical advantage to justify an $8.2 billion acquisition.

The Ethical and Regulatory Questions Grow With Physical AI

Spatial intelligence also creates challenges that go beyond computing performance.

When an AI system generates text, an incorrect output may create misinformation or a bad business decision. When an intelligent machine controls equipment, moves through a warehouse, manipulates infrastructure, or operates alongside people, mistakes can have physical consequences.

That raises the importance of robust evaluation, simulation, human oversight, cybersecurity, access controls, audit trails, and clear boundaries for autonomous action.

Privacy matters too.

Spatial AI systems may process rich visual representations of homes, workplaces, factories, streets, and other environments. Organizations will need clear rules around what data is captured, how long it is retained, how models learn from it, and who is allowed to access reconstructed environments.

Regulators are also likely to scrutinize major AI acquisitions from several angles, including competition, data governance, safety, and the concentration of foundational AI capabilities. The AMD–World Labs transaction itself remains subject to regulatory approval before it can close.

For enterprises, that means the physical-AI revolution should not be treated as purely an engineering challenge. Governance needs to develop alongside capability.

What Businesses Should Watch Next

The most important question is not whether AMD suddenly becomes a world-model company.

It is whether bringing model researchers and chip engineers under the same roof creates a meaningful engineering advantage.

Watch what happens to AMD’s future GPU and rack-scale designs. Watch how ROCm evolves for spatial and multimodal workloads. Watch whether Marble and future World Labs models become more tightly optimized for AMD accelerators. Watch what the company releases for robotics and simulation developers. And watch whether AMD’s promise of an open ecosystem translates into tools enterprises can actually deploy across heterogeneous infrastructure.

Most importantly, watch for signs that world models are moving from research projects into commercially useful platforms.

That transition is far from guaranteed. Even sophisticated world models still face hard problems involving physical accuracy, consistency, inference cost, reliability, training data, and the notorious gap between simulation and messy reality.

But an $8.2 billion acquisition suggests AMD believes those problems are worth attacking now.

Conclusion: AMD Is Betting on the Next Frontier of AI

AMD’s planned $8.2 billion acquisition of World Labs is more than a high-profile technology deal. It reflects a broader shift in artificial intelligence from systems that primarily understand text, images, and video toward models that can interpret, simulate, and eventually interact with the physical world.

By bringing Fei-Fei Li and the World Labs research team into AMD, the company is positioning itself closer to the frontier of spatial intelligence, world models, robotics, simulation, and physical AI. Just as importantly, AMD could gain valuable insight into how future AI workloads should shape the design of its GPUs, software platforms, networking technologies, and data center infrastructure.

The deal also highlights how rapidly the AI competitive landscape is evolving. Semiconductor companies are no longer competing only on chip performance. Increasingly, the race involves complete ecosystems that combine hardware, software, models, research talent, developer tools, and real-world applications.

There are still major questions ahead. AMD will need to successfully integrate World Labs, demonstrate practical commercial applications for spatial AI, maintain its commitment to an open ecosystem, and address the privacy, safety, governance, and regulatory challenges that come with AI systems capable of understanding physical environments.

Still, the strategic direction is clear. The next major phase of AI may be defined not simply by machines that can generate information, but by machines that can understand how the world works.

If AMD and World Labs can turn spatial intelligence into scalable, reliable, and commercially useful technology, this acquisition could become an important milestone in the evolution of AI computing.

Artificial intelligence in the enterprise is moving beyond answering questions and generating content. The next stage is considerably more ambitious: AI systems that can understand a business objective, interact with software, complete multiple steps, and take action across real operational workflows.

SAP is positioning itself directly in the middle of that transition.

At its Transformation Excellence Summit in Atlanta on September 22, 2026, SAP announced new capabilities across SAP Signavio, SAP LeanIX, WalkMe, and SAP Cloud ALM designed to connect AI agents more tightly with business processes, enterprise knowledge, employee workflows, governance, and ongoing operational monitoring. SAP describes these capabilities as part of the operational backbone for its broader vision of the autonomous enterprise.

The announcement is important because enterprise AI is reaching a point where intelligence alone is not enough. An agent that can reason impressively but does not understand company policy, authorization boundaries, process dependencies, or compliance requirements can quickly become more liability than productivity tool.

SAP’s strategy is therefore increasingly focused on a different challenge: giving AI agents enough business context to act usefully while maintaining enough oversight to keep those actions accountable.

For additional industry analysis of the announcement, Futurum’s detailed examination of SAP’s AI agent strategy provides useful context around process intelligence, agent ownership, risk monitoring, employee adoption, and return on investment. [SAP Links AI Agents to Business Processes and Oversight]

SAP Is Moving AI Agents Into Real Business Processes

The biggest change in enterprise AI is the move from assistance to execution.

Traditional generative AI might summarize a procurement report or explain why an invoice was rejected. An agentic system can potentially analyze the situation, gather additional information, interact with applications, initiate a workflow, route an approval, and monitor what happens next.

That changes the technology’s role dramatically.

SAP’s approach centers on integrating agents with the processes enterprises already depend on rather than treating AI as a disconnected chatbot floating above them.

The company’s SAP Signavio Process Consulting Agent, for example, can use process-mining capabilities across SAP and non-SAP environments to analyze performance, identify bottlenecks, and recommend areas where AI agents may create value. SAP Signavio Process Intelligence can then provide monitoring of how those processes actually perform.

SAP lays out this broader architecture in its own overview of the operational backbone of the autonomous enterprise, including the relationship among Signavio, LeanIX, WalkMe, and Cloud ALM. [SAP’s Opearational Backbone of the Autonomous Enterprise]

This is an important distinction. Successful enterprise agents will not simply need smarter models. They will need to understand how the company itself works.

Company Memory Gives AI Agents Business Context

One of SAP’s more interesting ideas is SAP Company Memory.

Businesses operate on far more than databases and process diagrams. Important operational knowledge is scattered across policies, documentation, workflows, standards, employee expertise, exceptions, and years of accumulated institutional experience.

An AI agent operating without this information might understand the technical task while still misunderstanding the business.

Company Memory is intended to create a governed source of organizational rules, process knowledge, standards, and operating context that can be used by both people and AI agents. SAP says the capability remains in preview.

This could be particularly valuable when an agent encounters situations where the technically possible action is not necessarily the organizationally acceptable one.

A procurement agent, for example, should not merely know how to create a purchase order. It may also need to understand approved vendors, spending thresholds, regional requirements, exception procedures, sustainability policies, contractual obligations, and which decisions still require human approval.

That institutional context is what can turn an impressive AI demo into something that can function inside an actual company.

SAP LeanIX Brings Governance Into the Agent Landscape

As companies deploy more agents, another problem appears quickly: agent sprawl.

An enterprise may eventually operate hundreds or thousands of agents from multiple vendors. Some may interact with HR data, others with financial records, customer systems, supply chains, analytics platforms, or external services.

Without centralized visibility, organizations may struggle to determine which agents exist, who owns them, what they can access, which models they use, and what risks they introduce.

SAP is addressing that challenge partly through SAP LeanIX and SAP AI Agent Hub.

SAP says AI Agent Hub can inventory agents, large language models, and Model Context Protocol servers, then associate those systems with specific business capabilities and owners. Its new AI Governance Assistant incorporates compliance intelligence related to frameworks including the EU AI Act and NIST guidance and can reassess classifications when systems change.

SAP has also published a broader discussion of agent sprawl and why AI governance is becoming an enterprise leadership issue, emphasizing inventory, ownership, permissions, monitoring, and lifecycle management

WalkMe Addresses the Often-Ignored Adoption Problem

Even perfectly governed AI is not particularly valuable when employees do not use it.

That is where WalkMe becomes strategically important to SAP’s plan.

Rather than requiring workers to leave their existing applications and visit a separate AI interface, WalkMe is designed to surface Joule Agents, assistants, and AI capabilities inside the workflows employees already use.

SAP is also extending agents into systems that may not provide modern APIs. WalkMe’s UI-native agent can operate through application interfaces, potentially allowing agents to interact with legacy applications, customized ERP systems, and third-party software.

Two additional capabilities support this approach. AI Authoring allows teams to describe a desired digital experience using natural language and generate a deployable plan, while AI Insights can analyze workflow and content performance and produce charts, dashboards, and related analysis.

This matters because enterprise transformation is rarely blocked solely by a lack of technology. Adoption, training, workflow design, organizational incentives, and employee trust frequently determine whether an AI project progresses beyond the pilot stage.

SAP Cloud ALM Extends Oversight After Deployment

Governance does not end when an AI agent goes live.

Production is where the interesting problems begin.

Models change. Applications change. permissions change. Business policies evolve. Data distributions shift. New integrations appear. An agent that behaved correctly during testing may encounter situations that were never represented in its original evaluation.

SAP Cloud ALM is intended to provide part of this operational layer.

SAP says Cloud ALM supports the transformation lifecycle through system analysis, configuration, testing, and rollout. After deployment, AI agent monitoring can track agent actions so operations teams can detect deviations, assess whether agents are behaving as expected, and measure business outcomes.

This represents an important evolution of AI governance.

Organizations cannot treat agent approval like installing conventional software and checking a compliance box. Agentic systems require continuous oversight.

That means monitoring actions, permissions, exceptions, performance, cost, risk, and business outcomes throughout the lifecycle.

ROI Will Decide Whether Enterprise Agents Scale

The agentic AI market currently has no shortage of exciting demonstrations. Enterprises, however, eventually need numbers.

Futurum’s September 2026 analysis reported that its Enterprise Software Decision Maker Survey of 833 respondents found expected and actual returns on recent software purchases averaging roughly 14% using its range-midpoint estimate. The same analysis reported that 41.5% of decision-makers said validated ROI studies would increase confidence in budget allocation.

Its CIO research also highlighted the gap between adoption and scaled returns: 73.3% of organizations buying AI externally reported either adoption without realized ROI or returns that remained limited to pilots, while 12.6% reported sustained ROI at scale.

Those numbers reinforce an increasingly important lesson.

The future of enterprise AI will be judged less by the number of agents a company deploys and more by whether those agents measurably improve business processes.

Useful metrics may include procurement cycle time, invoice-processing cost, days sales outstanding, employee time saved, customer-response time, error rates, exception volumes, and the total computing or token cost associated with agent execution.

The Bigger Shift Is From AI Tools to AI Operating Models

SAP’s latest announcement reveals where enterprise AI appears to be heading.

The first phase of generative AI focused heavily on models.

The second focused on copilots.

The next phase is increasingly about systems of agents working inside governed business processes.

That requires a much larger technology stack.

Agents need enterprise data. They need process context. They need permissions. They need organizational knowledge. They need application access. They need ownership. They need monitoring. They need audit trails. They need human escalation paths. And businesses need evidence that all of this actually improves performance.

SAP is attempting to connect those pieces through Signavio, LeanIX, WalkMe, Cloud ALM, Joule, and its broader Business AI platform.

The company is also emphasizing that governance cannot sit outside the agent architecture. In its discussion of controlling increasingly autonomous systems, SAP argues that governance, risk, and compliance mechanisms need to evolve alongside agents as they gain access to business processes and tools.

What Enterprise Leaders Should Take From SAP’s AI Agent Strategy

The most useful lesson from SAP’s approach is not limited to companies already running SAP.

Agentic AI changes the basic risk equation of enterprise software.

When AI only generates information, organizations primarily govern its outputs.

When AI can take action, organizations must also govern its identity, permissions, tools, processes, decisions, dependencies, and consequences.

That makes several priorities increasingly important for enterprise AI programs:

• Give agents access to governed business context rather than unrestricted information.
• Assign identifiable owners to agents and automated workflows.
• Apply least-privilege access controls.
• Maintain traceable logs of consequential actions.
• Define situations that require human review or approval.
• Monitor agents continuously after deployment.
• Train employees around new human-agent workflows.
• Measure business outcomes against clear pre-deployment baselines.
• Reassess risk whenever models, integrations, permissions, or processes change.

Organizations that treat AI agents simply as smarter chatbots may miss the real transformation.

The bigger shift is toward software that can participate directly in business operations.

Final Thoughts

SAP linking AI agents to business processes and oversight represents an important milestone in the evolution of enterprise AI.

The conversation is no longer only about whether an AI model can reason well enough to complete a benchmark or produce an impressive demo. Enterprises increasingly need to determine whether an agent understands their processes, follows their rules, respects authorization boundaries, works inside existing applications, leaves an audit trail, and produces measurable business value.

That is where SAP’s combination of Signavio process intelligence, Company Memory, LeanIX governance, WalkMe adoption tools, AI Agent Hub, Joule, and Cloud ALM becomes strategically interesting.

The race toward autonomous enterprise systems is accelerating, but autonomy without context and accountability is unlikely to be enough.

The companies that succeed with AI agents will be the ones that make the technology not only intelligent, but observable, governable, measurable, and useful inside the real-world processes that keep the business running.

Artificial intelligence has written code, generated images, predicted protein structures and helped researchers sift through mountains of scientific data. Now, a team of AI “scientists” has taken on something considerably more ambitious: helping identify a promising therapeutic strategy for lung cancer.

The headline sounds almost science-fictional. A virtual biotechnology company staffed by tens of thousands of AI agents examined clinical evidence, investigated biological targets and proposed a treatment approach involving the protein B7-H3, also known as CD276.

But the real story is more interesting than “AI invented a cancer drug.”

According to [Nature] Stanford researchers developed a system called Virtual Biotech, capable of coordinating as many as 37,000 specialized AI agents under an AI “chief scientific officer.” The system was designed to imitate the organizational structure of a real biotechnology company, with different agents handling areas such as target discovery, genomics, clinical-trial analysis and therapeutic development.

What emerged is an important glimpse of where AI-powered drug discovery may be headed: not toward one giant chatbot replacing scientists, but toward networks of specialized AI agents working alongside human researchers.

Meet the Virtual Biotech: A Drug Company Made of AI Agents

Drug discovery is not one problem. It is hundreds of interconnected problems.

Researchers need to understand disease biology, identify promising targets, examine genetic evidence, study cellular behavior, assess toxicity, select an appropriate drug design, interpret previous clinical trials and eventually determine how a new candidate should be tested.

That fragmentation is exactly what the Stanford-led team wanted to address.

The researchers describe the architecture in their [Virtual Biotech research platform] as an organization of specialized AI scientists coordinated by a virtual Chief Scientific Officer, or CSO. Instead of asking a single language model one enormous question, the CSO breaks a scientific problem into smaller assignments and delegates them to specialist agents equipped with appropriate datasets and analytical tools.

Think of it less like ChatGPT answering a biology question and more like a digital research organization holding a very, very crowded staff meeting.

The peer-reviewed research, published in Science, involved Harrison G. Zhang, Peter Eckmann, Jiacheng Miao, Andrew B. Mahon and James Zou. The [PubMed record for the Science paper] describes Virtual Biotech as a multi-agent framework designed to support therapeutic discovery and development across multiple scientific disciplines.

37,000 AI Agents and Nearly 56,000 Clinical Trials

Before tackling lung cancer, the researchers gave the system a monster-sized homework assignment.

Virtual Biotech analyzed outcomes from 55,984 clinical trials. More than 37,000 clinical-trialist agents participated in the analysis, with agents extracting structured information and connecting trial outcomes with genomic and cellular evidence.

The scale matters because pharmaceutical research generates a staggering volume of fragmented evidence. Clinical-trial registries, scientific papers, genomic databases and other research repositories often contain information that could affect drug-development decisions, but connecting those dots manually is slow.

According to [Stanford Medicine’s account of the project] the agents catalogued roughly 50,000 trials in less than a week—a job James Zou said could otherwise take human researchers years.

The agents then found an intriguing pattern.

Drugs aimed at genes whose activity was highly specific to particular cell types appeared to perform better than drugs with less specific targets. In the research dataset, drugs targeting these cell-type-specific genes were 40% more likely to progress from Phase I to Phase II, 48% more likely to reach market, and associated with 32% fewer adverse events.

Those percentages should be interpreted carefully. They represent associations found in historical data, not proof that targeting cell-specific genes automatically causes a drug to succeed.

But as a drug-development signal, the finding is valuable. If researchers can identify biological characteristics that correlate with better clinical outcomes, AI systems could help scientists prioritize promising targets before enormous amounts of time and money are committed.

How the AI Team Zeroed In on Lung Cancer

The researchers next asked Virtual Biotech to investigate B7-H3, a protein also known as CD276, as a possible therapeutic target in lung cancer.

B7-H3 was not an unknown protein magically discovered by the AI. Researchers were already studying it as an oncology target.

That distinction is important.

What Virtual Biotech did was assemble and analyze evidence from multiple sources to assess why the target might be useful and what therapeutic approach could make sense.

According to Stanford Medicine, the agents found evidence that B7-H3 was highly expressed in fibroblasts—connective-tissue cells that can exist around tumors. Further analyses suggested that B7-H3-expressing fibroblasts could participate in signaling interactions that suppress nearby immune cells, potentially helping cancer hide from immune defenses.

The AI agents then recommended an antibody-drug conjugate, or ADC.

If you are not spending your evenings reading oncology journals—and we will forgive you—an ADC can be thought of as a precision delivery vehicle. An antibody recognizes a protein found on particular cells, while a potent cancer-killing payload is attached to it. The objective is to bring that payload more directly to cells expressing the target instead of distributing chemotherapy as broadly throughout the body.

The Virtual Biotech agents proposed a B7-H3-directed ADC strategy. [Phys. org’s coverage of the study] offers another useful overview of how the agents arrived at the treatment approach and how researchers plan to test additional AI-generated hypotheses in physical laboratories.

The Next Challenge: Getting AI Out of the Computer and Into the Lab

There is one rather stubborn problem with computational drug discovery: biology does not care how impressive your demo looks.

A hypothesis eventually needs to survive contact with the physical world.

Cells must be tested. Experiments need to be reproduced. Toxicity has to be evaluated. Animal studies may be required. Manufacturing processes need development. Human clinical trials must establish safety and efficacy. Regulators then have to evaluate the evidence.

Nature’s coverage emphasizes this limitation: outside researchers noted that Virtual Biotech itself has not yet been validated through the complete real-world drug-development process and that many of its predictions still require experimental testing.

The researchers make essentially the same point. The [bioRxiv version of the Virtual Biotech research] presents the system as a framework for generating and evaluating therapeutic hypotheses—not a substitute for experimental validation.

AI Drug Discovery Also Creates New Governance Questions

The more autonomy we give scientific AI systems, the more important accountability becomes.

What happens when an AI agent cites a paper incorrectly? How should a pharmaceutical company audit a conclusion generated by thousands of interacting agents? Who is responsible for catching subtle biological assumptions? How do regulators evaluate research where important decisions were partly generated by AI?

These are not reasons to abandon the technology. They are reasons to build validation, transparency and human review into the technology from the beginning.

Agentic scientific systems will need traceable evidence, reproducible analyses, strong data governance and clear boundaries around which decisions require human authorization.

This is particularly important in healthcare, where “the model sounded confident” is not a scientific validation strategy.

What This Breakthrough Really Means

The most fascinating thing about Virtual Biotech is not the number 37,000.

After all, spinning up thousands of software agents is much easier than hiring thousands of scientists.

The important development is coordination.

For years, AI systems have excelled at narrowly defined scientific tasks: predicting structures, classifying images, searching chemical space or summarizing papers. Virtual Biotech points toward systems that can combine many of those capabilities into a coordinated research process.

That could change the economics of early-stage drug development.

Instead of researchers manually spending months navigating fragmented datasets and literature, AI agents could continuously evaluate evidence and surface promising hypotheses for experts to investigate. Human scientists could spend more time designing experiments, challenging assumptions and making high-level scientific judgments.

The result would not be “AI replacing scientists.”

It could be scientists suddenly gaining access to a research organization that can expand or shrink on demand.

And that may be the real breakthrough.

Conclusion

The idea of thousands of AI agents working together to uncover promising cancer treatments may sound futuristic, but it is quickly becoming part of real scientific research. Stanford’s Virtual Biotech shows what can happen when AI moves beyond answering questions and starts coordinating complex research tasks across clinical trials, biology, and drug development.

This does not mean AI has replaced scientists—or that it has independently cured lung cancer. What it does show is that AI can help researchers process enormous amounts of information faster, connect evidence that might otherwise be missed, and generate stronger hypotheses for human experts to test.

The biggest opportunity may be collaboration. As AI systems become more specialized, transparent, and connected to real laboratory workflows, scientists could gain access to powerful digital research teams that accelerate discovery without removing the need for human judgment, validation, and oversight.

For biotech, pharmaceutical companies, and healthcare innovators, this is a glimpse of what comes next: AI not just as a tool, but as a research partner helping turn massive datasets into meaningful scientific progress.

Artificial intelligence is transforming healthcare, but access to advanced medical technology remains uneven. While physicians at well-funded hospitals can consult extensive medical databases, specialist networks, and sophisticated diagnostic tools, healthcare professionals in underserved regions often face limited access to the same resources.

A newly announced partnership between Anthropic and OpenEvidence aims to narrow that gap.

The two companies announced a collaboration to expand access to AI-powered clinical decision support in low- and middle-income countries. By combining Anthropic’s AI technology with OpenEvidence’s medical knowledge platform, the initiative seeks to help healthcare professionals access reliable medical research and treatment guidance at no cost.

The collaboration raises an important question for the healthcare industry: Could making evidence-based medical AI more accessible help physicians deliver better-informed care, regardless of their location?

Here is what the partnership involves, how the technology works, and what healthcare organizations need to understand about its opportunities, limitations, and responsible implementation.

What Is the Anthropic–OpenEvidence Partnership?

The Anthropic–OpenEvidence partnership is a global healthcare AI initiative designed to make advanced medical knowledge more accessible to physicians, particularly those working in resource-constrained healthcare environments.

According to [Reuters’ report on the Anthropic–OpenEvidence partnership] the collaboration will introduce a specialized version of OpenEvidence that healthcare professionals in dozens of low- and middle-income countries can access for free.

The partnership combines the capabilities of two organizations working in complementary areas of artificial intelligence.

Anthropic: The AI technology provider

Anthropic develops Claude, a family of advanced AI models designed to support tasks such as reasoning, research, information analysis, and complex knowledge work.

Within this partnership, Anthropic will provide the underlying AI technology that supports the medical knowledge platform.

Anthropic’s healthcare ambitions extend beyond this collaboration. In its official blog post, [Advancing Claude in Healthcare and the Life Sciences] the company explains how it has expanded Claude’s healthcare capabilities, including tools designed for healthcare providers, health technology companies, and life sciences organizations.

These developments illustrate how advanced AI models are increasingly being adapted for specialized healthcare applications rather than functioning exclusively as general-purpose chatbots.

OpenEvidence: The medical knowledge specialist

OpenEvidence is an AI-powered medical knowledge platform designed to help healthcare professionals find relevant clinical information.

Rather than requiring clinicians to manually search through multiple medical journals, the platform allows them to ask clinical questions and receive responses informed by medical research and treatment guidelines.

Why Expanding Medical AI Worldwide Matters

Access to quality healthcare is not determined by medical expertise alone. It also depends on infrastructure, technology, financial resources, and the availability of reliable information.

Consider two physicians treating patients with similar conditions.

One works at a major medical institution with access to specialized medical databases, experienced colleagues, and extensive research resources.

The other works in a rural healthcare facility where specialist consultations may be difficult to arrange, access to medical literature is limited, and essential diagnostic equipment may not always be available.

Both physicians need accurate medical information, but they do not necessarily have equal access to it.

This is the gap the Anthropic–OpenEvidence partnership aims to address.

Bringing medical knowledge closer to underserved communities

AI-powered clinical decision support could help clinicians navigate relevant medical research without requiring expensive institutional subscriptions to multiple information services.

Instead of searching several databases independently, a physician could use a conversational interface to explore treatment guidelines, investigate possible explanations for a patient’s symptoms, or review research relevant to a particular clinical question.

For healthcare professionals working without immediate access to specialists, this could provide an additional source of information to support their decision-making.

The potential benefits extend beyond individual consultations. Easier access to medical research could support continuing medical education, help clinicians keep up with changing guidance, and encourage more consistent use of evidence-based practices.

Why accessibility must go beyond affordability

Making an AI platform free is an important step, but affordability is only one component of meaningful healthcare accessibility.

Healthcare organizations also need reliable connectivity, appropriate devices, digital literacy, training, and access to the medicines and diagnostic equipment required to act on medical recommendations.

A sophisticated AI system offers limited practical value if it recommends a diagnostic test that is unavailable locally or a medication that patients cannot obtain.

This makes the partnership’s emphasis on adapting the platform to local healthcare conditions especially important.

The broader objective of equitable AI access is also reflected in [Anthropic’s blog post about its $200 million partnership with the Gates Foundation]

Announced in May 2026, that separate initiative includes funding, AI usage credits, and technical support for projects involving global health, life sciences, education, and economic mobility. Its health-related work includes exploring how AI can support frontline healthcare workers and healthcare decision-making in low- and middle-income countries.

The OpenEvidence collaboration is another example of how advanced AI is being explored as a tool for making healthcare knowledge more widely available.

How Anthropic and OpenEvidence Could Transform Clinical Decision Support

Clinical decision support refers to tools that help healthcare professionals interpret information and make informed decisions about patient care.

Traditional clinical decision support systems frequently rely on predefined rules, alerts, medical databases, or structured recommendations.

Generative AI introduces a more flexible interface through which clinicians can ask questions in natural language and receive synthesized information.

The Anthropic–OpenEvidence partnership brings these capabilities together.

Faster access to evidence-based medical information

Medical research continues to expand, making it difficult for busy physicians to follow every relevant development across multiple specialties.

An AI-powered medical knowledge platform can potentially help reduce the time clinicians spend locating and reviewing relevant information.

For example, a physician might ask for current evidence concerning the management of a particular condition in a patient with multiple existing health problems.

A clinical decision support system could help identify relevant guidelines, summarize research findings, and highlight considerations that warrant further investigation.

The clinician would still need to examine the supporting evidence and determine whether the information applies to the individual patient’s circumstances.

This distinction is essential: retrieving relevant medical information is not the same as independently making a diagnosis or choosing treatment.

Turning medical research into practical clinical knowledge

Clinical decisions often require information from multiple sources, including research studies, professional guidelines, and established treatment protocols.

The challenge is not simply finding an article. It is understanding what the evidence means for a particular situation.

AI can potentially assist by organizing information into a more accessible format, highlighting differences between research findings, and making it easier for clinicians to investigate specific questions.

Anthropic’s [Claude Science: An AI Workbench for Scientists] provides additional context for this broader technological direction.

The June 2026 announcement describes a research environment designed to help scientists work with scientific databases, computational tools, and auditable research outputs.

Although Claude Science and the Anthropic–OpenEvidence clinical platform are distinct initiatives, they demonstrate how AI is evolving toward specialized systems that help professionals work with complex scientific information.

For the medical community, the important opportunity is to make trustworthy information easier to find and evaluate without weakening the scientific standards that guide patient care.

Why Local Healthcare Context Is Central to the Partnership

One of the most significant aspects of the Anthropic–OpenEvidence announcement is its emphasis on adapting medical AI to different healthcare environments.

A treatment recommendation that is appropriate in a well-equipped hospital may be difficult to implement in a healthcare facility with limited diagnostic resources.

Likewise, medical guidance developed primarily for one population may not fully reflect the disease patterns, treatment availability, or clinical circumstances found elsewhere.

Adapting AI to regional medical needs

Reuters reports that OpenEvidence has been working with healthcare organizations in Rwanda and Botswana to adapt its technology to local medical conditions and available resources.

These efforts illustrate why global healthcare AI cannot simply be a standardized medical chatbot distributed internationally.

A locally relevant platform needs to account for factors such as disease prevalence, treatment availability, clinical infrastructure, and applicable medical guidelines.

It should also recognize when available evidence does not adequately address a particular population or clinical situation.

For example, a physician working in a facility without advanced imaging equipment may need information about diagnostic approaches that are practical within the facility’s existing resources.

A useful clinical AI platform should help the physician understand relevant options and their limitations rather than assume that every healthcare facility operates under identical conditions.

The role of smartphones in expanding access

Mobile accessibility could be another important factor in making medical AI more widely available.

Smartphones may provide healthcare professionals with access to clinical information even when their institutions lack extensive desktop computing resources.

However, smartphone availability does not eliminate problems involving network reliability, electricity, device affordability, or the secure handling of patient information.

For this initiative to deliver sustainable value, participating healthcare organizations will need to consider how the technology performs under actual operating conditions.

The opportunity is substantial, but deployment must be designed around the healthcare environments in which clinicians work.

The Biggest Challenges: Accuracy, Privacy, and Responsible AI

Expanding access to medical AI is a promising development, but healthcare technology must meet higher standards of reliability than many other AI applications.

An inaccurate business summary might lead to an inconvenient decision. Incorrect medical information could have much more serious consequences.

The long-term value of the Anthropic–OpenEvidence partnership will therefore depend on how effectively the technology addresses clinical accuracy, patient privacy, algorithmic bias, and appropriate human oversight.

Clinical accuracy and AI-generated misinformation

Generative AI systems can produce convincing responses that contain inaccurate, incomplete, or unsupported information.

Even when a medical AI platform draws on peer-reviewed research, clinicians must be able to assess whether its conclusions accurately reflect the original evidence.

This means a reliable clinical decision support system should provide transparent references, communicate uncertainty, and distinguish established medical guidance from emerging research.

Patient privacy and data protection

Medical information is among the most sensitive categories of personal data.

Healthcare organizations adopting AI-powered tools must carefully evaluate how patient information is collected, processed, stored, and shared.

Important considerations include data access permissions, encryption, consent, retention policies, and compliance with applicable healthcare privacy requirements.

The specific responsibilities of each organization will depend on the platform’s technical architecture, the information being processed, and the legal requirements of the country in which the technology is deployed.

Healthcare providers should not assume that a platform’s availability automatically makes it appropriate for entering identifiable patient information.

Algorithmic bias and healthcare inequalities

Medical AI systems may perform differently across patient populations when the underlying evidence, training data, or evaluation methods do not adequately represent those populations.

This risk becomes especially important when technology developed in one healthcare environment is deployed internationally.

A medical AI platform should be evaluated for its performance across relevant geographic, demographic, and clinical settings.

Local healthcare professionals should also participate in reviewing recommendations, identifying inappropriate assumptions, and assessing whether the information reflects regional standards of care.

What the Partnership Means for Healthcare Technology Companies

The Anthropic–OpenEvidence partnership offers valuable insights for healthcare technology companies, software developers, hospital administrators, and organizations considering AI-powered clinical applications.

It demonstrates that healthcare AI innovation is increasingly focused on developing specialized tools for specific professional needs.

Rather than building a generic chatbot and positioning it as a medical expert, technology companies can design solutions around carefully defined healthcare workflows.

For example, an AI-powered medical knowledge platform might be developed to support literature retrieval, clinical guideline navigation, or medical education.

Each application has distinct requirements for accuracy, security, usability, and clinical validation.

Building healthcare AI around real-world needs

Organizations exploring similar solutions should begin by identifying the specific problems they want to address.

Is the objective to reduce the time required to find medical information? Improve access to continuing education? Help clinicians navigate treatment guidelines? Or support more efficient clinical documentation?

The answer should determine the application’s design, intended users, information sources, and evaluation criteria.

Developers should also consider practical deployment requirements, including mobile compatibility, accessibility, system integration, and the technical capabilities of participating healthcare facilities.

Google’s article, [How Google Is Using AI to Improve Health for Everyone] offers another perspective on the broader healthcare AI landscape.

Published in March 2026, the post discusses initiatives involving clinical AI education, rural healthcare accessibility, and tools intended to help people access and understand health information.

Together, these developments highlight the growing importance of designing AI solutions around healthcare accessibility, professional expertise, and measurable practical value.

Understanding healthcare AI regulations

Healthcare AI developers must also consider whether their applications are subject to medical device regulations.

Regulatory requirements differ by country and depend on the software’s intended purpose, functionality, and clinical use.

The Future of Medical AI: From Information Access to Better Healthcare Delivery

The Anthropic–OpenEvidence partnership reflects a broader shift in the artificial intelligence industry.

Advanced AI is increasingly being developed for specialized scientific, medical, and professional applications rather than remaining limited to general-purpose conversational tools.

The healthcare sector is particularly relevant because clinicians frequently need to interpret complex information under time constraints.

However, the future of medical AI will depend on more than model capabilities.

Healthcare organizations will need evidence that these tools improve clinical workflows, provide accurate and relevant information, and support better decision-making without introducing unacceptable risks.

The partnership also raises several questions that will become important as its international deployment progresses.

How will the platform be evaluated across different countries? What safeguards will protect patient information? How will local medical professionals participate in adapting the technology? And how will participating organizations measure its impact on clinical practice?

These are important questions for any company developing medical AI solutions.

The growing role of artificial intelligence in medicine also extends beyond clinical decision support.

AI is increasingly being explored in pharmaceutical research, genetic analysis, scientific literature review, and drug development.

Conclusion: Making Medical AI Accessible Is an Important Step Toward More Equitable Healthcare

The Anthropic–OpenEvidence partnership represents an important development in the international expansion of artificial intelligence for healthcare.

By combining advanced AI technology with evidence-based medical knowledge, the initiative aims to make clinical decision support more accessible to healthcare professionals working in underserved regions.

Its emphasis on free access and local healthcare adaptation addresses two important barriers to global AI adoption: affordability and practical relevance.

However, expanding access is only the beginning.

The initiative’s long-term impact will depend on whether the technology consistently provides accurate medical information, protects sensitive data, reflects local clinical conditions, and integrates effectively into established healthcare workflows.

For healthcare organizations and technology developers, the underlying lesson is clear: successful medical AI adoption requires more than powerful algorithms. It demands trustworthy information, responsible governance, clinical expertise, and technology designed around the needs of the people using it.

The future of healthcare AI is not simply about making medical information available everywhere. It is about helping healthcare professionals turn reliable information into better-informed decisions, wherever they practice.

Novo Nordisk—now increasingly using the shortened name Novo—announced a collaboration with Anthropic focused on applying Claude AI to pharmaceutical research and development. The objective is ambitious but practical: help researchers solve difficult scientific problems, improve biological reasoning, accelerate software development, and ultimately shorten the path between an early research question and a potential new medicine. Novo’s official announcement says the initial effort will involve testing Claude Science on specific R&D workflows and identifying scientific problems where the combined expertise of Novo and Anthropic could have the greatest impact. [Novo]

That distinction matters. Anthropic and Novo have not announced that Claude is independently inventing a particular new drug or replacing experimental scientists. What they are building is arguably more interesting: an AI-assisted research environment capable of connecting scientific literature, data, computational tools, reasoning, software development, and human expertise.

Reuters independently reported the collaboration as part of Novo’s broader effort to speed medicine development using AI. [Reuters]

Claude Science Could Become an AI Workbench for Novo Researchers

The most important technology in this collaboration may be Claude Science, Anthropic’s specialized AI environment for scientific research.

This is not simply the regular Claude chatbot with a collection of biology prompts attached. Anthropic designed Claude Science to integrate many of the tools researchers already use, manage computational workloads, analyze scientific literature and data, create reproducible research artifacts, and coordinate specialist AI agents.

According to Anthropic, Claude Science includes more than 60 curated skills and connectors across areas such as genomics, proteomics, structural biology, single-cell analysis and cheminformatics. It can also interact with scientific databases and computational resources while preserving an audit trail showing how analyses, figures and conclusions were produced. [Anthropic]

For a pharmaceutical company, that could be significant.

Drug discovery rarely involves one database, one experiment or one brilliant eureka moment. Scientists may need to evaluate published research, examine genomic data, compare biological pathways, analyze molecular properties, run computational models, inspect previous experimental results and decide what experiment should happen next.

Traditionally, much of that work happens across fragmented systems.

Claude Science is designed to act as something closer to an intelligent coordination layer across those systems.

Claude Could Help Researchers Reason Across Complex Biological Evidence

Novo says the partnership will specifically explore workflows that support biological reasoning.

That phrase deserves attention.

Biology is extraordinarily interconnected. Understanding whether a potential therapeutic target is worth pursuing can involve genetics, protein interactions, cellular pathways, disease mechanisms, experimental data, existing literature and evidence from previous programs.

Humans are exceptionally good at forming hypotheses and applying scientific judgment. Humans are considerably less good at simultaneously reading millions of documents and databases.

That is where an advanced AI research system becomes useful.

Claude could potentially help researchers gather evidence from different sources, identify relationships between findings, summarize contradictory evidence, compare hypotheses and expose gaps that deserve further investigation. Scientists can then evaluate those outputs and determine which ideas deserve experimental testing.

Chemistry Is Another Area Where Claude Is Becoming More Capable

Anthropic has also been working directly on Claude’s ability to reason about chemistry.

In June 2026, the company published research exploring Claude’s performance on nuclear magnetic resonance, or NMR, spectroscopy—a fundamental technique chemists use to determine molecular structures.

Anthropic tested Claude models against established chemistry tools on a set of compounds from research published after the models’ training cutoff. The company reported particularly strong results from its most capable model on parts of the NMR prediction task, while also emphasizing that the broader goal is to assist chemists with translation, analysis and information integration rather than eliminate expert judgment.

Why does that matter for the Novo partnership?

Drug research constantly moves between different representations of scientific information: molecular structures, journal figures, experimental measurements, database entries, chemical notation and laboratory results.

An AI system capable of understanding several of these formats could reduce repetitive analytical work and help scientists move between them more efficiently.

Instead of spending valuable research hours translating information between systems, researchers could spend more time asking the questions that actually require human scientific expertise.

Agentic AI Could Change How Pharmaceutical R&D Software Is Built

The partnership is not limited to biology.

Novo specifically says it intends to use Anthropic’s frontier models to strengthen AI-driven software development, describing software engineering as an important enabler for scaling AI across the company.

That may sound less exciting than discovering a new molecule, but it could be just as important operationally.

Modern pharmaceutical R&D runs on enormous amounts of software. Bioinformatics pipelines, research databases, analytical dashboards, simulation environments, clinical systems, laboratory software and internal scientific platforms all need to be built, maintained and connected.

Agentic coding systems can potentially help developers create prototypes, write and test code, integrate systems, automate repetitive engineering work and maintain research applications.

The implication is that Claude could influence Novo’s R&D from two directions.

On one side, Claude Science can assist the scientist.

On the other, Claude’s coding and agentic capabilities can assist the engineers building the infrastructure those scientists depend on.

That combination may be more strategically important than deploying a single headline-grabbing “AI drug discovery model.”

The Next Frontier Could Connect Claude to Physical Laboratory Work

There is another development worth watching.

In August 2026, Anthropic introduced its Model Hardware Standard, or MHS, a research-preview framework designed to let AI agents communicate with programmable physical equipment such as microscopes, liquid handlers and robotic arms.

That opens the door to a longer-term possibility where an AI research system does not only analyze experimental results. It could potentially coordinate parts of the experimental workflow itself.

Imagine an agent reviewing existing evidence, recommending an experiment, coordinating compatible laboratory equipment, examining the results and helping researchers determine what should be tested next.

That is not what Novo and Anthropic have publicly promised in their new collaboration. But it demonstrates the direction in which Anthropic’s scientific AI ecosystem is evolving.

Responsible AI Will Be as Important as Faster AI

Healthcare and pharmaceutical research are unforgiving environments for unreliable automation.

A hallucinated restaurant recommendation is annoying.

A fabricated scientific citation, incorrect calculation or poorly understood biological inference entering an R&D decision can be considerably more serious.

That makes auditability, validation and human oversight central to the Anthropic–Novo collaboration.

Novo explicitly states that the collaboration has been designed around robust data governance and human oversight so that AI is used in accordance with its ethical and compliance standards.

Claude Science is also designed around reproducibility. Anthropic says generated scientific artifacts can include the underlying code, computational environment, explanations and message history, while a reviewer agent can inspect citations, calculations and relationships between figures and their underlying code.

That is particularly relevant in pharmaceutical research, where a clever answer is not enough. Teams need to understand where information came from, what assumptions were made, how an analysis was performed and whether another researcher can reproduce it.

Conclusion: Claude AI Could Reshape How Drug Discovery Gets Done

The Anthropic–Novo collaboration reflects a broader shift in pharmaceutical research: AI is moving beyond simple productivity tools and becoming part of the scientific workflow itself. By combining Claude’s reasoning, coding, and scientific research capabilities with Novo’s pharmaceutical expertise, proprietary data, and established R&D processes, the partnership could help researchers analyze evidence faster, improve biological reasoning, streamline software development, and identify promising research directions more efficiently.

The most important point, however, is that Claude AI for drug discovery is not about replacing scientists. Its value will depend on how effectively it augments human expertise while maintaining rigorous validation, transparency, reproducibility, data governance, and regulatory oversight. In drug development, speed matters—but trustworthy science matters more.

If Anthropic and Novo can demonstrate measurable improvements in research productivity and decision-making without compromising scientific standards, their work could provide a model for how generative and agentic AI are adopted across the pharmaceutical industry.

For technology leaders, researchers, and healthcare organizations, this is a development worth following closely. The next major breakthrough in AI-powered drug discovery may not come from a machine independently inventing a medicine. It may come from giving scientists a far more capable set of tools for asking better questions, connecting more evidence, testing ideas faster, and turning promising discoveries into real-world treatments.

There are roughly three billion DNA base pairs in the human genome. At each position, the existing DNA letter can theoretically be replaced by one of the other three letters.

Do the math and you arrive at a rather intimidating number: about nine billion possible single-letter changes.

Testing every one of those changes experimentally would be practically impossible. Google DeepMind’s answer is to let artificial intelligence do the first pass.

In September 2026, Google DeepMind introduced AlphaGenome Atlas, a massive AI-generated catalogue containing predictions for the molecular effects of approximately nine billion single-nucleotide variants—essentially every possible one-letter substitution across the human genome. Google describes it as its most comprehensive effort yet to map how genetic mutations could alter molecular biology. Readers looking for the technical announcement can explore [Google DeepMind’s AlphaGenome Atlas launch article]

It is an ambitious idea, even by modern AI standards. Instead of asking an AI model about one genetic variant at a time, DeepMind has essentially asked it the genomic equivalent of: “What happens if we change every letter, one by one?”

What Is Google’s AlphaGenome Atlas?

AlphaGenome Atlas builds on AlphaGenome, a genomic AI model that Google DeepMind first unveiled in 2025.

AlphaGenome was designed to analyze extremely long DNA sequences—up to one million base pairs—and predict thousands of molecular characteristics associated with gene regulation. These include effects involving gene expression, RNA splicing, chromatin accessibility and other mechanisms that influence how genetic information is actually used by cells. For a deeper look at the underlying model, Google DeepMind’s earlier [AlphaGenome introduction explains how the system approaches regulatory variant prediction].

That distinction matters.

When most people hear the word gene, they naturally think about DNA that directly encodes proteins. Yet only around 2% of the human genome is protein-coding. Much of the remaining genome participates in controlling when, where and how genes are activated, while other regions have functions that remain poorly understood.

Think of protein-coding DNA as the recipes in a cookbook. Non-coding regulatory DNA contains a huge amount of the information telling the kitchen which recipe to use, when to use it, in what quantity and under what conditions.

A one-letter change in the wrong regulatory location can therefore matter even when it does not directly alter a protein.

That difficult territory is where AlphaGenome becomes particularly interesting.

Why Nine Billion DNA Variants?

The calculation behind the headline is surprisingly straightforward.

The human reference genome contains roughly three billion DNA positions. DNA uses four nucleotide bases—A, C, G and T. At any given position, the letter already present could theoretically be replaced by any of the other three.

Three billion positions multiplied by three possible substitutions gives approximately nine billion single-nucleotide variants.

AlphaGenome Atlas precomputes predictions for those substitutions rather than forcing researchers to repeatedly run the model themselves.

That could turn an expensive computational question into something closer to a database lookup.

As [Nature] individual nucleotide substitutions are among the most common forms of human genetic variation. Some have little or no meaningful effect, while others can contribute to disease risk or, in rare cases, play a direct role in genetic disease.

The challenge is figuring out which is which.

Nine billion possibilities make “check them all in the lab” a fairly terrible project plan.

AI offers a way to prioritize.

The AlphaGenome Variant Impact Score Could Be Just as Important as the Atlas

Producing billions of predictions creates another problem: researchers still need a practical way to navigate them.

Google DeepMind therefore introduced the AlphaGenome Variant Impact, or AVI, score.

The AVI score combines information from AlphaGenome with predictions from AlphaMissense, DeepMind’s system for assessing protein-altering variants. The goal is to provide researchers with a single metric for prioritizing potentially consequential variants across both coding and non-coding regions.

The Atlas can also break the predicted impact down into biological categories, helping researchers investigate why a particular variant receives attention—for example, whether the model predicts disruption to RNA splicing, gene expression or chromatin-related processes.

This is important because useful scientific AI cannot simply shout, “Interesting mutation!” and walk out of the room.

Why AlphaGenome Could Matter for Rare Disease Research

One particularly promising application is variant prioritization in rare diseases.

Genome sequencing can reveal enormous numbers of differences between an individual genome and a reference genome. The hard part is often determining which differences are biologically meaningful.

AlphaGenome Atlas could help researchers rank candidate variants before committing scarce laboratory resources to experimental validation.

Google says collaborators have already used the Atlas to investigate rare-disease variants. One reported example involved researchers using its AVI score to prioritize a variant affecting the DNM1 gene, with AlphaGenome predicting the creation of an incorrect splice site.

That does not mean AI has “diagnosed” the disease. It means AI has helped researchers narrow the search space and generate a mechanism worth investigating experimentally.

AI Is Especially Useful in the Genome’s “Control Panel”

The most intriguing part of AlphaGenome may be its ability to interrogate non-coding DNA.

Traditional genetics understandably focused heavily on protein-coding genes because their biological effects were easier to connect to specific mechanisms. Regulatory DNA is messier.

Its influence can depend on cell type, surrounding sequence, chromatin state and complex interactions with proteins and other molecular machinery.

AI models are particularly attractive for this type of problem because they can search enormous datasets for context-dependent patterns that are difficult to express through simple rules.

But there is an important catch.

[Ars Technica] notes that AlphaGenome is currently based on the biological datasets and cell types available to researchers. Predictions involving poorly characterized biological contexts therefore remain an area where real-world validation will matter enormously.

In other words, AI can offer researchers a very good map.

It does not guarantee that every road on that map has been driven.

AlphaGenome Is a Research Tool, Not a Medical Diagnosis Machine

This may be the most important point for anyone reading about AlphaGenome outside the genomics community.

AlphaGenome Atlas is not a clinical diagnostic system.

Google DeepMind explicitly states that information produced by AlphaGenome Atlas is not a substitute for professional medical advice, diagnosis or treatment and that the system has not been validated or approved for clinical use.

Predicting a molecular effect is not the same thing as predicting whether a person will develop a disease.

Human health emerges from complicated interactions among multiple genes, regulatory processes, environment, lifestyle, development, age and many other factors.

This is why responsible AI becomes particularly important once artificial intelligence moves from web search and productivity tools into biology and medicine.

Genomic AI Creates New Ethical Questions Too

The better AI becomes at interpreting genomes, the more important governance becomes.

Genetic information is unusually sensitive because it can reveal information not only about one person but also, indirectly, about biological relatives. Organizations combining genomic datasets with increasingly powerful AI models therefore need strong controls covering consent, access, security, data provenance and appropriate downstream use.

Bias deserves equal attention.

Genomic datasets do not represent every population equally. A model that performs strongly against widely studied populations, tissues or cell types may not automatically generalize equally well everywhere else.

Scientific teams should therefore treat AI predictions as evidence to investigate—not infallible biological truth.

Conclusion

Google’s AlphaGenome Atlas shows how powerful AI is becoming in the study of human genetics. By evaluating billions of possible one-letter DNA changes, the system can help researchers quickly identify which genetic variants may be worth studying more closely.

This does not mean AI can replace doctors, geneticists, or laboratory research. Instead, tools like AlphaGenome can help scientists narrow down huge amounts of genetic information, develop better research questions, and focus experiments on the most promising possibilities.

The real value of Google’s AI genome system is speed and scale. What could take researchers an enormous amount of time to investigate manually can now be explored much more efficiently with AI. As these tools improve, they could support advances in rare disease research, drug discovery, and our broader understanding of how the human genome works.

At the same time, accuracy, privacy, fairness, and responsible use will remain essential. AI can point researchers in the right direction, but human expertise and scientific testing are still needed to confirm what those predictions actually mean.

Ultimately, AlphaGenome Atlas is another sign that AI is becoming an important partner in scientific discovery—helping researchers turn billions of possibilities into useful questions that could lead to better insights about human health.

Artificial intelligence has spent years getting better at thinking inside computers. Now one of its most important challenges is learning how to work outside them.

That transition—from software that generates answers to machines that can perceive, reason and physically act—is driving the rapidly expanding field of Physical AI. And a new European partnership suggests that the race is moving beyond impressive robot demonstrations toward something much harder: industrial-scale production.

Germany-based NEURA Robotics and Italian embedded-computing specialist SECO announced a strategic partnership designed to industrialize and scale Physical AI from Europe. Under the agreement, SECO will help design, engineer and manufacture electronic compute modules for NEURA’s cognitive robots, including its humanoid 4NE1. The systems will incorporate Qualcomm Dragonwing processors as part of NEURA’s distributed computing architecture. The full partnership details are available in the [Seco] 

That might initially sound like a hardware supply agreement.

It is considerably more interesting than that.

The partnership links robotics, edge AI computing, electronics manufacturing, industrial data and semiconductor production into what could become a broader European Physical AI ecosystem. And it illustrates something increasingly important about the robotics market: building an intelligent machine is no longer just a robotics problem. It is a full-stack computing, data and manufacturing problem.

The NEURA–SECO Partnership Is About Turning Robotics Into an Industry

Robotics companies are very good at producing videos that make the future look approximately three weeks away.

Commercial-scale manufacturing is less cinematic.

A robot intended to operate reliably in a factory needs industrial-grade computing, sensing, power management, safety systems, real-time control, networking, thermal engineering, durable electronics and a supply chain capable of producing the same system repeatedly.

That is where SECO enters the picture.

SECO specializes in embedded computing, edge AI and industrial electronics. Under the partnership, the company will engineer and manufacture compute modules for NEURA robots, helping move those systems from advanced platforms toward repeatable series production in Europe. Independent coverage from [Intelligent CIO] highlights the same industrialization angle: NEURA contributes cognitive robotics and Physical AI technology, while SECO provides the engineering and manufacturing capability required to turn that intelligence into scalable hardware.

That distinction matters.

There is a huge difference between building 20 sophisticated robots and reliably producing thousands of them. Manufacturing scale introduces entirely new constraints involving component availability, quality assurance, power efficiency, maintainability and cost.

The NEURA Robotics and SECO Physical AI partnership is therefore partly an attempt to solve the decidedly unglamorous problems that separate a great prototype from a viable industry.

And those problems may ultimately determine which robotics companies survive.

NEURA’s “Brain + Nervous System” Approach Could Be the More Important Story

Traditional robots often resemble centralized machines: sensor information travels to one principal computing system, decisions are made, and commands travel back to the motors and actuators.

NEURA is pursuing a more distributed architecture.

The company describes its approach as a “Brain + Nervous System” architecture, supported by its Smart Limb concept. Instead of concentrating every computational decision in one place, sensing and processing can be distributed throughout the robot and positioned closer to the joints, sensors and mechanisms where physical actions occur.

Think less “one computer operating a robot” and more “an intelligent network spread throughout the machine.”

There is a practical reason for that approach.

A cognitive robot may be reasoning at several different timescales simultaneously. High-level intelligence might determine that a component needs to be picked up and moved. Local systems must then handle much faster tasks such as maintaining balance, interpreting force, controlling a joint or reacting immediately when unexpected contact occurs.

Smart Limbs Bring AI Closer to Physical Action

Smart Limbs are particularly interesting because they point toward a different way of thinking about robot intelligence.

Imagine a worker catching a falling cup.

The brain does not need to consciously calculate every muscle movement from scratch. Different parts of the nervous system coordinate perception and physical response extremely quickly.

Robotic systems obviously work very differently from biological bodies, but the architectural principle offers a useful analogy. Distributed intelligence can allow individual robot subsystems to handle tightly constrained real-time tasks while a higher-level system concentrates on more complex reasoning.

That creates several potential advantages.

It can reduce latency, limit the amount of sensor information constantly traveling to a central processor, support modular hardware designs and improve responsiveness when robots operate close to people.

Qualcomm’s [Qualcomm] shows how seriously the broader robotics industry is taking this hardware challenge. The reference platform integrates computing, sensing, networking and software with the explicit goal of helping developers move from prototypes toward deployable robots. NEURA is listed among the ecosystem participants supporting the platform.

The Semiconductor Industry Could Become a Training Ground for Physical AI

The partnership becomes even more strategically interesting when you look beyond NEURA’s own robots.

NEURA and SECO plan to collaborate on Physical AI systems for semiconductor and electronics manufacturing, using real-world production environments to generate industrial data and develop new automation capabilities.

That creates a potentially powerful feedback loop.

Robots operate inside real production environments. Their interactions produce data. That data can help developers improve perception, manipulation and decision-making. New skills can then be refined, validated and deployed across additional robots and manufacturing sites.

The objective is not simply to program one machine to repeat one movement forever.

It is to create reusable intelligence.

Coverage from [Unite] emphasizes this connection between European manufacturing, robot compute modules and scalable Physical AI infrastructure.

This is also why real-world industrial data is becoming so valuable.

Generative AI could learn enormous amounts from material already stored on the internet. Physical AI does not have the same luxury. A humanoid learning to manipulate irregular components, respond to changing factory conditions or recover from unsuccessful actions needs data describing physical interactions—not just text describing them.

NEURA Gym Could Help Build the Data Flywheel

NEURA and SECO are also discussing the creation of a NEURA Gym in Italy, which would become NEURA’s first training hub in Southern Europe.

The name may sound like a place where humanoids finally get serious about leg day, but the concept addresses one of robotics’ biggest bottlenecks: training.

Physical AI systems need experience.

Developers require environments where robots can practice tasks, collect interaction data, validate new capabilities and improve models before those systems enter production environments.

Real-world training hubs can complement simulation by providing data involving friction, imperfect components, unexpected obstacles, human behavior and other details that digital environments may struggle to reproduce perfectly.

That creates what the AI industry increasingly describes as a data flywheel: deploy systems, observe what happens, learn from those interactions, improve the models and redeploy the improvements.

NEURA’s broader Neuraverse vision extends the idea further. Skills developed through one robot or application could eventually become reusable capabilities available to other machines rather than remaining isolated inside a single custom automation project.

This shift from isolated automation toward reusable physical intelligence could be one of the most commercially important developments in robotics.

Why Europe Matters to This Partnership

The words “from Europe” in the announcement are not decorative.

The partnership is explicitly framed around developing and manufacturing important parts of the Physical AI stack in Europe.

Europe already possesses several ingredients that could make it unusually competitive in Physical AI: world-class manufacturing, industrial automation, automotive engineering, robotics research, sensors, embedded systems and enormous quantities of specialized industrial knowledge.

Where Europe has struggled more visibly is in the enormous centralized computing infrastructure and frontier AI platforms that have made American technology companies dominant during the generative-AI boom.

Physical AI changes the competitive equation somewhat.

A robot in an automotive factory or semiconductor facility requires more than a giant language model. It needs mechanical engineering, embedded electronics, real-time systems, industrial data, sensors, manufacturing capabilities, safety expertise and close integration with real production workflows.

What Businesses Should Watch Next

For executives, manufacturers and technology leaders, the significance of the NEURA–SECO agreement extends beyond whether 4NE1 becomes the dominant humanoid robot.

The partnership reveals several signals worth watching:

• Physical AI is moving toward industrialization. The conversation is increasingly about manufacturing capacity, embedded computing and repeatable deployment rather than demonstrations alone.
• Edge computing will be critical. Robots need local intelligence capable of responding quickly even when cloud connectivity is limited.
• Real-world operational data may become a competitive asset. Companies that can responsibly turn production experience into reusable robotic skills could create powerful feedback loops.
• Ecosystems may matter more than individual vendors. Robotics increasingly requires collaboration among AI developers, semiconductor companies, manufacturers, integrators and industrial customers.
• Europe has a plausible Physical AI advantage. Its industrial base, robotics expertise and engineering infrastructure could allow it to compete differently from regions dominating cloud-based generative AI.

That last point deserves particular attention.

The Physical AI race may not be won by the company capable of producing the most entertaining humanoid demonstration.

It may be won by the ecosystem that can manufacture robots reliably, train them on valuable real-world tasks, update them safely and repeat that process thousands of times.

Conclusion

The partnership between NEURA Robotics and SECO shows how quickly Physical AI is moving from research and prototypes into real-world manufacturing. By combining NEURA’s intelligent robots with SECO’s embedded computing and production expertise, the two companies are working to make advanced robotics easier to build, scale, and deploy across Europe.

The bigger story is that AI is no longer limited to software on a screen. It is becoming part of machines that can see, understand, move, and work alongside people. For businesses, this could mean smarter factories, more flexible automation, and new ways to improve productivity.

Europe also has an opportunity to play a major role in this shift because of its strong manufacturing, robotics, and engineering industries. If partnerships like NEURA and SECO succeed, Europe could become an important center for the development of safe, reliable, and scalable Physical AI.

Ultimately, the future of AI may not only be about systems that can answer questions. It may also be about intelligent machines that can understand the physical world and take useful action within it.

Artificial intelligence is beginning to change the economics of the global technology services industry, and Wipro has just provided one of the clearest examples yet of how significant that change could become.

[Reuters] the Indian IT services company says its artificial intelligence initiatives have generated productivity improvements equivalent to the work of approximately 20,000 employees. Importantly, Wipro says this does not mean 20,000 people were simply replaced. Instead, employees whose capacity was freed by AI have been redeployed to other projects, trained for different roles, or tasked with managing AI-powered systems.

With Wipro employing roughly 243,000 people as of June 2026, the productivity capacity being discussed is equivalent to about 8% of its workforce. That makes Wipro AI productivity an important case study in how enterprise AI may reshape work without necessarily translating automation directly into one-for-one job elimination.

Wipro Is Moving Toward a Human-AI Operating Model

Wipro CTO Sandhya Arun described the company’s direction as a move toward a “human-AI operating model.” Rather than separating employees and artificial intelligence into competing groups, this model treats AI agents as tools or digital workers that employees can supervise, coordinate and use to increase their own output.

An engineer who once handled a collection of repetitive development or support tasks manually, for example, could increasingly oversee multiple AI agents performing portions of that work. The employee can then spend more time on architecture, client requirements, quality control, decision-making and other higher-value activities.

The strategy did not appear overnight. Wipro launched its [Wipro] in 2023 alongside a commitment to invest $1 billion in artificial intelligence capabilities over three years. The initiative was designed to integrate AI across Wipro’s internal operations, platforms, solutions, research and client services while simultaneously expanding AI training across the workforce.

The latest productivity figures suggest those investments are now moving beyond experimentation and beginning to influence how work is actually delivered.

AI Training Is Becoming as Important as AI Technology

Perhaps the most important part of Wipro’s strategy is not the AI itself but what the company is doing with its people.

Reuters reports that more than 100,000 Wipro employees have received advanced AI-related training and certifications as the company expands its human-AI model.

That matters because the next phase of enterprise AI is likely to require more than knowing how to use a chatbot. Employees may increasingly need to understand how to supervise agents, validate AI-generated output, design AI-enabled workflows, connect models with enterprise systems and determine when human intervention is necessary.

This aligns with [IBM] IBM argues that businesses seeing the greatest value from AI are increasingly redesigning how their organizations operate rather than simply deploying more AI tools.

AI Is Also Changing the IT Services Business Model

Wipro’s productivity gains matter far beyond its own workforce.

India’s enormous technology services sector has traditionally relied heavily on people-based delivery models. In simple terms, large teams of engineers and consultants perform work for clients, while contracts frequently reflect the amount of talent and time required to deliver it.

AI challenges that structure.

A separate [Reuters] found that major providers including Wipro, Tata Consultancy Services, Infosys, HCLTech and Cognizant are increasingly facing customers who expect faster delivery and greater productivity for lower costs. Contracts are consequently moving toward business outcomes rather than simply hours worked.

That shift creates both an opportunity and a challenge.

If AI allows an IT provider to complete the same project faster with fewer manual hours, productivity improves. But clients also know AI makes those efficiencies possible, meaning they may demand a share of the savings.

The competitive advantage therefore cannot simply be, “We use AI.”

Almost every major technology services company can make that claim.

The stronger advantage will come from demonstrating that AI can produce measurable improvements in areas such as customer experience, software delivery speed, revenue generation, service quality and operating costs.

That is why Arun’s emphasis on moving from productivity to outcomes may ultimately be more significant than the headline 20,000-worker-equivalent figure.

Forward-Deployed Engineers Could Become a Critical AI Role

Another emerging piece of Wipro’s strategy involves forward-deployed engineers.

These specialists typically work closely with customers to take AI from a demonstration or prototype into an operational system. Instead of simply developing technology from a distance, they embed more deeply with client teams, understand domain-specific problems and adapt AI systems to real business environments.

Wipro is expanding this capability, following a broader movement across India’s technology services industry.

This could become increasingly valuable because enterprise AI deployment is rarely as simple as connecting a large language model to company data. Businesses must consider security, permissions, legacy technology, governance, workflow design, compliance and human oversight.

[McKinsey] similarly argues that organizations need to rethink hiring, internal capabilities and workforce structures as AI agents take on greater portions of technology work.

In other words, AI may reduce demand for certain repetitive tasks while increasing demand for professionals who can successfully turn AI capabilities into working business systems.

The Next Test for Wipro: Turning AI Productivity Into Revenue

There is still an important unanswered question.

Productivity gains demonstrate that AI can improve internal efficiency, but businesses ultimately need to convert those efficiencies into sustainable financial value.

Reuters notes that Wipro remains the only company among India’s four largest IT services providers that does not disclose AI revenue separately. Analysts cited by Reuters say the company is still relatively early in translating its AI investments into commercial returns compared with some competitors.

That makes the next phase particularly important.

If Wipro can combine AI-enabled productivity with stronger client outcomes, larger AI projects and improved margins, the 20,000-worker-equivalent capacity figure may eventually look like an early milestone rather than the main story.

If those productivity improvements primarily result in clients demanding lower prices, however, the financial impact could be more complicated.

What Wipro’s AI Push Means for Business Leaders

Wipro’s experience provides a useful blueprint for companies considering their own AI transformation.

The biggest opportunity may not come from simply reducing headcount. Instead, organizations can use AI to remove repetitive work, increase employee capacity and redirect human expertise toward higher-value problems.

That requires investment in technology, but it also requires training, workflow redesign, governance and clearly defined business outcomes.

AI adoption is increasingly becoming an organizational transformation challenge rather than purely a technology challenge.

The companies that succeed may be those that learn how to combine the speed and scalability of artificial intelligence with human judgment, domain expertise, creativity and accountability.

Conclusion

Wipro’s AI push shows that the biggest impact of artificial intelligence may not be simple job replacement, but a major shift in how work is organized. By freeing capacity equivalent to around 20,000 workers, the company is demonstrating how AI can handle repetitive tasks, improve productivity, and give employees more room to focus on higher-value work.

The real test now is whether Wipro can turn those efficiency gains into stronger client outcomes, better services, and measurable business growth. For other companies, the lesson is clear: successful AI adoption requires more than deploying new tools. It also depends on reskilling employees, redesigning workflows, and creating a model where people and AI work together effectively.

As enterprise AI continues to mature, businesses that focus on this balance between automation and human expertise may be best positioned to gain long-term value from the technology.

The age of the AI chatbot may be giving way to something considerably more powerful—and considerably more complicated.

Meta introduced Muse, a personal AI agent designed not merely to answer questions but to take actions on a user’s behalf. The Meta AI agent can connect with other apps and services, send emails, book travel, fill out forms, make purchases and continue working on tasks even after a user closes the app.

That is a substantial leap from asking an AI, “Can you draft this email?” to saying, “Handle this,” and watching the software actually do it.

According to [Meta’s official Muse announcement] the company sees Muse as part of a broader shift toward personal AI capable of turning goals into actions rather than simply producing information.

The launch puts Meta squarely into the rapidly expanding race to build agentic AI—AI systems that can reason about a goal, interact with outside software and perform a sequence of actions with less step-by-step human involvement.

And while the productivity possibilities are huge, so are the questions around permissions, privacy, security, financial transactions and accountability.

What Exactly Is Meta Muse?

Muse is Meta’s attempt to move AI from an assistant sitting inside a chat window to something closer to a digital operator.

Traditional generative AI generally waits for a question and generates an answer. An AI agent can instead interpret a goal, create a plan, use tools and execute parts of that plan.

Meta says Muse runs using its agent-focused AI technology and operates within a dedicated cloud-based environment designed to let the system interact with connected services while maintaining security controls.

That environment gives Muse access to a browser-like workspace for completing tasks across different services. The system can also continue working in the background while the user does something else.

What Can the Meta AI Agent Actually Do?

Meta is positioning Muse as a general-purpose personal agent rather than an assistant dedicated to one application.

The company says users can connect services across areas including email, calendars, payments, shopping, dining, entertainment and other digital services. From there, the agent can coordinate actions spanning multiple applications.

Muse can potentially:

• Send emails and manage communication
• Research and book travel
• Fill out online forms
• Make purchases
• Build shopping and grocery lists
• Coordinate calendars and invitations
• Track longer-term goals and projects
• Remember user preferences
• Continue certain tasks in the background
• Work across supported third-party apps rather than remaining inside one Meta product

The significance is not any one of those features. The important part is cross-application execution.

The [Associated Press coverage of Meta’s personal AI agent] highlights the growing difference between traditional conversational AI and agentic systems capable of acting across digital services.

The Privacy Question: How Much Should Your AI Know?

Muse becomes more useful as it understands more about the user.

That is simultaneously its biggest advantage and one of its biggest privacy challenges.

A personal agent might benefit from knowing your calendar, travel preferences, favorite stores, dietary requirements, purchasing habits and communication patterns.

But bringing those pieces together can potentially create an unusually detailed representation of an individual’s digital life.

Meta has emphasized privacy controls as part of the Muse design. The company outlines its approach to user data, connected services and future privacy protections in its [official product announcement].

The challenge is that personal agents require context to become genuinely useful.

An assistant that knows nothing about you cannot do much on your behalf.

An assistant that knows everything about you raises a very different set of questions.

Finding the right balance may become one of the most important product-design problems in consumer AI.

Security Concerns Haven’t Disappeared

Meta has made security one of the central themes of the Muse launch. However, independent reporting also shows why caution remains appropriate.

[Reuters’ reporting on the Muse launch] examined Meta’s development process and reported concerns arising during internal testing, including reliability and data-handling issues.

That is arguably the central tension surrounding every powerful AI agent.

An agent needs enough authority to be useful.

Every additional permission also expands the consequences when the agent misunderstands instructions, encounters malicious content, receives a deceptive prompt or simply fails.

[TechCrunch’s analysis of Muse and the consumer trust question] also focuses on a key challenge facing Meta: convincing consumers to give an AI system access to services containing valuable and highly personal information.

The technical problem is therefore only half the challenge.

The other half is earning enough trust that people will actually connect those accounts.

Why Meta Muse Matters for Businesses

Muse is launching as a consumer-focused product, but enterprise technology leaders should pay close attention.

Consumer platforms often create expectations that eventually migrate into the workplace.

Employees accustomed to telling a personal AI, “Plan this trip and book it,” will increasingly expect business software to respond to requests such as:

“Review these invoices, identify discrepancies, update the system and send me the exceptions.”

Or:

“Find available meeting times across these teams, prepare the briefing material and schedule the meeting.”

The underlying shift is from software as a collection of interfaces to AI as an orchestration layer across software.

That could significantly affect SaaS platforms, ecommerce companies, financial institutions, CRM providers, travel businesses and virtually any company whose customers currently interact through websites and apps.

Tomorrow’s customer may not always be a human clicking your interface.

Sometimes it may be an agent operating on the human’s behalf.

Is Muse the Future of Personal AI?

Possibly—but the most important part of Muse is not whether this particular product becomes the dominant personal agent.

It is what the launch tells us about where the technology industry is heading.

Meta is betting that the next phase of AI is not primarily about answering better questions.

It is about taking better actions.

The company has enormous distribution through its existing platforms, giving it a potential path to put agentic AI in front of mainstream consumers rather than limiting the technology to developers and early adopters.

[Reuters’ coverage] provides additional detail on Meta’s positioning of Muse and the broader competitive race around personal AI agents.

If agents become reliable enough, the familiar routine of opening apps, navigating menus and manually transferring information between services could gradually disappear.

But the winners will not simply be the companies whose agents can perform the greatest number of tasks.

They will be the companies whose agents users trust enough to let them perform those tasks.

Conclusion: Meta Muse Signals the Next Phase of AI

Meta’s Muse AI agent is a clear sign that artificial intelligence is moving beyond conversation and into action. The ability to access apps, send emails, coordinate tasks and participate in payments could make AI far more useful in everyday life—but it also raises the stakes for privacy, security, permissions and accountability.

For businesses, the takeaway is bigger than one Meta product. Agentic AI is becoming a new interface for how people may interact with software, services and commerce. Companies that prepare now for AI-driven workflows, secure integrations and human approval systems will be better positioned as this shift accelerates.

For consumers, the key question will be trust. The most successful AI agents will not simply be the ones that can do the most—they will be the ones people feel comfortable allowing into their inboxes, calendars, accounts and financial workflows.

Meta Muse may still be an early step, but it points toward a future where AI does more than recommend the next move. It may increasingly make that move for us.