Tag Archive for: Requirements & Requirements Management Page 5
Tag Archive for: Requirements & Requirements Management
Navigating FDA AI Guidance for Medical Devices: A Practical Guide
For medical device professionals, the integration of Artificial Intelligence (AI) and Machine Learning (ML) represents a monumental leap forward in innovation. However, this progress comes with significant regulatory hurdles. As AI algorithms evolve, so do the rules that govern them, leaving many development, quality, and regulatory teams struggling to keep pace. Failing to understand and adapt to the latest FDA AI guidance can lead to submission delays, compliance issues, and costly rework.
This guide delivers a practical overview of the evolving FDA regulatory framework for AI and ML-based medical devices, drawing on both recent draft guidance and the agency’s longer-term action plans. We highlight essential concepts including the Predetermined Change Control Plan (PCCP), Good Machine Learning Practices (GMLP), and Real-World Performance (RWP) monitoring and show how these shape the compliance landscape for manufacturers.
TL;DR: The FDA is moving toward a holistic Total Product Lifecycle (TPLC) regulatory approach for AI/ML-enabled medical devices, emphasizing continuous monitoring, clear GMLP, and mechanisms for pre-planned algorithm updates. Robust, traceable documentation, and proactive lifecycle risk management are now essential for compliance and product success.
The FDA’s Evolving AI/ML Regulatory Framework
The FDA has signaled its commitment to adapting device oversight in response to rapid advances in AI/ML. Traditionally, regulatory submissions were point-in-time events. Now, regulators recognize that adaptive, learning systems require ongoing oversight, especially as software “learns” from real-world experience.
Key foundational documents illustrate this evolution:
FDA’s 2021 AI/ML-Based Software as a Medical Device (SaMD) Action Plan: This action plan lays out five pillars to modernize oversight including development of a tailored regulatory framework, advancement of GMLP, fostering transparency with users, promoting methodologies for bias/robustness, and supporting real-world performance pilots.
Artificial Intelligence-Enabled Device Software Functions: Lifecycle Management and Marketing Submission Recommendations (Draft Guidance, 2025): This draft guidance details expectations for managing AI within medical devices throughout the entire product lifecycle, including design, labeling, bias mitigation, cybersecurity, postmarket surveillance, and the importance of the Predetermined Change Control Plan.
Clinical Decision Support Software Guidance (2026): Clarifies FDA’s criteria for Clinical Decision Support (CDS) software functions, offering practical examples to distinguish between Non-Device CDS such as software functions excluded from device regulation and those that remain under device oversight.
FDA AI/ML-Enabled Medical Devices List: Provides a current catalog of FDA-authorized devices using AI/ML technologies, helping manufacturers benchmark their projects and understand regulatory precedent.
In summary: The FDA’s approach now encompasses both initial submissions and ongoing, risk-based management, aligning regulatory expectations with the unique characteristics of AI/ML-driven technologies.
Introduced in both the 2021 action plan and expanded in the draft 2025 guidance, a PCCP enables manufacturers to define anticipated modifications to an AI/ML algorithm upfront. The plan specifies “what” may be changed (pre-specifications) and “how” changes are managed (an algorithm change protocol). This approach recognizes the evolving nature of AI/ML models, especially those learning from real-world use.
2. Good Machine Learning Practices (GMLP)
The FDA calls for GMLP, which are best practices covering data management, training procedures, documentation, interpretability, and bias mitigation, all aligned with consensus standards. GMLP underpins both product quality and regulator confidence, reducing the risk of unexpected outcomes or patient harm (See Action Plan Pillar 2).
3. Transparency and User Trust
Both guidance documents emphasize transparency for end users including clinicians, patients, and caregivers. Clear labeling, robust documentation, and transparency about model logic, data sources, and limitations are expected to build trust in AI/ML-powered devices.
4. Real-World Performance (RWP) Monitoring
Unlike static software devices, AI/ML-based products must demonstrate ongoing safety and efficacy. The FDA encourages collection and review of real-world data as part of postmarket surveillance. Manufacturers should implement plans for ongoing performance monitoring by adapting both processes and documentation to ensure device quality over time.
5. Bias Mitigation and Robustness
AI/ML algorithms can inadvertently encode biases from historical datasets. The FDA expects proactive identification and management of bias through diverse, representative training data, ongoing performance validation, and transparent reporting on limitations and subgroup analysis.
Your design history, risk management, GMLP adherence, model versions, data sets, and algorithm updates should all be auditable and linked. Use digital solutions for traceability and compliance, making audit preparation seamless.
Step 3: Prepare and Maintain a PCCP
If your product uses adaptive algorithms, develop a comprehensive Predetermined Change Control Plan. Detail the types of future modifications, associated risk controls, and your process for validating postmarket changes.
Step 4: Embrace Ongoing RWP Monitoring
Postmarket surveillance now means real-world performance tracking including collecting user feedback, monitoring for data drift, bias, and managing field updates in a proactive, traceable way.
Step 5: Differentiate Wellness from Medical Claims
Consult the Wellness Policy to determine if any features of your device are exempt from device regulation and document your rationale.
Frequently Asked Questions
Q: What’s the difference between Software as a Medical Device (SaMD) and AI in Medical Devices (AiMD)? A: SaMD refers to software that is itself a medical device. AiMD is software that is integrated into a physical device. Both fall under the FDA’s AI/ML regulatory frameworks.
Q: Is a PCCP mandatory for all AI-enabled devices? A: PCCPs are expected for devices with adaptive/evolving algorithms. Rigid, non-learning AI products may not need a PCCP, but processes for documenting and justifying updates are still required (draft guidance, 2025).
Q: How should we implement GMLP? A: Follow best practices outlined by the FDA and consensus standards. Ensure your team manages data, training processes, versioning, and labeling in a repeatable, controlled, and demonstrable manner.
Master the Complexity of AI Medical Device Development
The regulatory landscape for AI medical devices is complex, but it shouldn’t stifle innovation. By adopting an integrated approach with a live digital thread, you can manage the intricate web of requirements, risks, and data that define modern device development. This not only prepares you to pass audits with confidence but also empowers your teams to build safer, more effective products faster.
Jama Connect®, enhanced with AI-powered features in Jama Connect Advisor™, provides the end-to-end traceability needed to manage the development of complex AI-enabled systems. Streamline your documentation, automate traceability, and ensure your team is always audit-ready.
Note: This article was drafted with the aid of AI. Additional content, edits for accuracy, and industry expertise by Tom Rish.
In this blog, we’ll recap a section of our recent Expert Perspectives video, “A Method to Assess Benefit-Risk More Objectively for Healthcare Applications” – Click HERE to watch it in it entirety.
Expert Perspectives: A Method to Assess Benefit-Risk More Objectively for Healthcare Applications
Welcome to our Expert Perspectives Series, where we showcase insights from leading experts in complex product, systems, and software development. Covering industries from medical devices to aerospace and defense, we feature thought leaders who are shaping the future of their fields.
Assessing benefit‑risk is a foundational requirement for medical device manufacturers, yet it has long been one of the most challenging aspects of risk management. While risks are analyzed with rigor and precision, benefits are often described qualitatively, making objective comparisons difficult and slowing decision‑making across the product lifecycle.
A new, revolutionary method for assessing benefit‑risk changes that dynamic by unifying benefit and risk into a single, objective framework. Our expert perspectives video, “A Method to Assess Benefit-Risk More Objectively for Healthcare Applications,” offers actionable insights for healthcare innovators aiming to meet rigorous regulatory requirements while ensuring patient safety and efficacy.
In this episode of Expert Perspectives, Richard Matt explains how his method, dubbed the “Grand Unified Theory of Risk Management”, enables medical device companies to perform benefit-risk analyses with unprecedented speed and precision, delivering definitive determinations within minutes. This efficiency allows for multiple assessments throughout a project, unlocking opportunities to refine patient populations, expand product indications, and even use a benefit-risk assessment as a design parameter during development. Beyond product development, this method also provides a robust framework for addressing regulatory requirements, post-market analysis, and quality management system evaluations.
By transforming a traditionally subjective process into a data-driven, objective methodology, Richard Matt’s approach empowers healthcare innovators to bring safer, more effective solutions to market. For a deeper dive into this method and its implications, download the whitepaper from Aspen Medical Risk Consulting.
Below is a preview of our interview. Click HERE to watch it in its entirety.
Kenzie Jonsson: Welcome to our expert perspective series where we showcase insights from leading experts in complex product, systems, and software development. Covering industries from medical devices to aerospace and defense, we feature thought leaders who are shaping the future in their fields. I’m Kenzie, your host, and today, I’m excited to welcome Richard Matt. Formerly educated in mechanical, electrical, and software engineering and mathematics, Richard has more than thirty years of experience in product development and product remediation. Richard has worked with everyone from Honeywell to Pfizer and is now a renowned risk management consultant. Today, Richard will be speaking with us about his patent pending method to assess benefit-risk more objectively in health care. Without further ado, I’d like to welcome Richard Matt.
Richard Matt: Hello. My name is Richard Matt, and I’m delighted to be speaking with you about our general solution to the problem of assessing whether the benefit of a medical action will outweigh its risk. I’ll start my presentation by saying a few words about my background and how this background led to the benefit-risk method you’ll be seeing in the presentation.
To understand my background, it really helps to go back to the first job I got out of undergraduate school. I graduated with a degree in mechanical engineering and an emphasis in fluid flow. And my first job was in the aerospace industry at Arnold Engineering Development Center, at a wind tunnel that Baron von Braun designed. I worked there as a project manager, coordinating various departments with the needs of a client who brought models to be tested. These are pictures of the ADC’s transonic wind tunnel with its twenty-foot by forty-foot long test section that consumes over a quarter million horsepower when running flat out. Those dots in the walls are holes, and a slight suction would pull the out on the outside of the wall to suck the air’s boundary layer through the holes. So a flight vehicle appeared more closely to match its flight air characteristics in free air. It was amazing place to work.
We could talk about aerodynamic issues and thermodynamic issues like why nitrogen condenses out of the air at mach speeds above six or why every jet fighter in every country’s air force has a maximum speed of about mach three and a half. But to stay on the topic of benefit-risk, the reason or my intro to this, the reason I was brought this up was that I saw here firsthand the long looping iterations that came from different technical specialties, each approaching the same problem from the respective of their technical specialty. I found it very frustrating and the, following analogy very apt, after getting, so each of our technical specialties would look at the same problem, the elephant from their own view. And I found myself getting frustrated with my electrical and software engineering coworkers, that they didn’t understand what I was talking about, but I knew realized soon I didn’t understand what they were talking about either.
So I decided I wanted to become part of the solution to that problem by going back to graduate school and getting myself rounded out and my education so I could talk to these folks from their perspective also. So I went back to grad after mechanical and undergraduate, went back to graduate school in electrical and mathematics and picked up enough software. I started teaching, programming also in college. I developed there a solution for the robot arms in those wind tunnels to to control a robot arm for every possible one, two, or three rotational degree of freedom arm, and that was my graduate thesis. After I completed my thesis, I felt empowered to start, my work doing going wherever I wanted doing whatever I wanted to do and realized that if I wanted to do anything significant, it would take many years, and I decided to focus on teamwork. Does that sound pretty good?
Matt: My ability to work across technical boundaries enabled me to bring exceptional products to the market. For instance, I brought an Internet of Things (IoT) device to the market during the 1990s before Internet of Things was a thing. My leadership in product development advanced rapidly, culminating in as a VP of Engineering at a boutique design firm in the Silicon Valley.
And, the combination of the breadth of my formal training and my system perspective for solving problems has really helped me work across continue to work across boundaries, so that I’ve worked for companies to help them establish their pro product requirements, trace requirements, do V and V work. I’ve done a lot of post-market surveillance work. I established internal audit programs. I’ve been the lead auditee when my firm is audited. Done had significant success accelerating product development and has been on work on. So mixed in with all of these works, I special I started specializing into risk management as consulting focus versus something I just did normally during development.
And since the defense of a patent requires notice, I’ll mention that the material here is being pursued on the patent, and, would like to talk with anyone who finds this interesting to pursue after you’ve learned about it. So let me start my presentation on benefit risk analysis by talking about how important it is to all branches of medicine and the many problems we have implementing it. The solution I’m gonna come up with, I’ll just outline here briefly so you can follow as we’re going through the presentation. I’m gonna first establish a single and much more objective metric to measure benefit and risk than people traditionally use. I’ll be accumulating overall benefit and risk with sets of metric values from this first metric. And finally, we’ll show how to draw a conclusion from the overall benefits and risk measurements of which is bigger benefit or risk.
So in terms of importance, historically, benefit-risk has been with medicine for millennia. It’s a basic tenant to all of medicine. The first do no harm goes all the way back to the quarter of Hammurabi 2,000 BC, and it legally required physicians to think not just about how they can help patients with treatment or what harm they might cause to treatment and making sure that the balance of those two favor the patient is very much the benefit-risk balance that we look at today. The result we’re gonna talk about is gonna be used everywhere throughout medicine with devices, with drugs, with biologics, even with clinical trials.
So is that fundamental cross medicine? How it’s used currently?
If you are in one of the ways developing new products, benefit-risk determinations have to be used in clinical trials to show that they’re ethical to perform, that we’re not putting people in danger needlessly. Benefit-risk determinations are the final gate before a new product is released for use to patients. And I have a quote here from a paper put out by AstraZeneca saying the benefit-risk determination is the Apex deliverable of any r and d organization. There’s a lot of truth to that. It’s the final thing that’s being put together to justify a product’s release. And so it has a very important role here for FDA and has a very important role for pretty much the regulatory structure of every country, including the EU.
Matt: In terms of creating a quality system, every medical company is required to have one. Benefit-risk determinations are used to assess a company’s quality system. This is per the FDA notice about factors on benefit-risk analysis. When regulators are evaluating company’s quality system, they’ll use benefit-risk to determine if nothing should be done, if a product should be redesigned, if they should take legal actions against a company of a range of possibilities from replacing things in the field to stopping products from being shipped. It’s also a key in favorite target for product liability lawsuits, because of how subjective it is, and we’ll get to that in a moment. It can also be used for legal actions against officers. So benefit risk is a really foundational concept for getting products out and keeping products out and keeping companies running well. Just a bit of historical perspective of medical documentation and development. We have here, I cited four different provisions of the laws, regarding medical devices in the United States. This is a small sampling.
The point I’m trying to make here is that each of these summaries of the laws discuss continually evolving, continually growing, more rigorous standards for evidence, more detailed requests for information from the regulators to the instrumentation development companies to the product development companies. So first, medical products are heavily regulated. We have the trend of increasing analysis and rigor. Per ISO 142471, and this is an application standard that is highly respected in the medical device field. A decision as to whether risks are outweighed with benefits is essentially a matter of judgment by experienced and knowledgeable individuals.
And this is our current state of the art.
Not that everybody does it this way, but this is the most common method of performing benefit-risk analysis. And benefit-risk analysis by this method, has a lot of problems because it’s based on the judgment and it’s based on individuals, and both of those can change with different settings. That’s why it’s a favorite point of attack for product liability lawsuits.
This quote was true in 1976, when medical devices were put under FDA regulation, but significantly remains unchanged nearly fifty years laters. Benefit-risk determinations are an aberration and that unlike the rest of medicine, they have not improved over time. They’ve remained a judgment by a group of individuals. In, twenty eighteen, FDA was, approached by congress to set a goal for itself of increasing the clarity, transparency, and consistency of benefit risk assessments from the FDA.
This was in human drug review as the subject, and the issue was that various drug companies had gotten very frustrated with the FDA for disagreeing with their assessments of what benefit-risk should look like. And to repeat again, when you have a group of individuals making a judgment, that’s gonna lead to inconsistencies because both the group and their own individual judgment will vary from one situation to the next. I have another, quote here from the article from AstraZeneca. The field of formal and structured benefit-risk assessments is relatively new.
Matt: Over the last twenty years, there’s still a lack of consistent operating detail in terms of best practice by sponsors and health authorities. So this is an understatement, but a true statement. We have had a lot of increasing effort over the last few years because if people are dissatisfied with the state of benefit-risk assessments, they want to do better than this judgment approach. And so there have been a plethora of new methods developed. I’ve found one survey here that summarize fifty different methods just to give you an idea of how many attempts there are. And I went through those fifty methods.
The other thing that’s interesting to see is the FDA’s attempt to clarify benefit-risk assessments. I have here five guidance documents from the FTA, and I would put forth the proposition that anytime you need five temps five attempts to explain something, it means you didn’t understand the thing well in the first place or failing about a bit trying to get it done right. I think this is also held up by the drug companies, pressure on congress to get FDA to improve their clarity and consistency of benefit-risk assessments.
So here’s the, fifty methods that I found in one study of benefit-risk assessments. They have them grouped into, a framework, metrics, estimate techniques, and utility surveys. These are the fifty different methods, and I’ve gone through each one of them. And they all have fundamental problems. They, I’m going through them a bit slowly. Like, here’s one, from the FDA, another benefit risk assessment. Health-adjusted life years are one of the few that uses the same metric for benefit and risk. Number needed to treat is a very popular indication for a single characteristic, but you can’t integrate that across the many factors that needed to do benefit-risk assessment.
And so we’ve gone down the rest of these, methods. If I group these fifty methods by how they accumulate risk, I get a rather useful collection. Most of the methods do not consider all the risk-benefit factors for benefit-risk situation. They will pick on just one factor. And you can’t combine the factors with themselves or with others. It’s simply looking at one factor by itself. So it’s an extremely narrow view of benefit-risk for most of these. The few methods that do look at all the risk-benefit factors, most of them start with what I call the judgment method, where you’re forced to distill all the factors down to the most significant few, only four maybe four to seven methods, four to seven factors.
So either the methods consider only one type of, one factor at a time, or they force you to throw away most of the methods and consider maybe four or seven factors is the second method. The third method is they assign numbers to the factors, they’ll add the factors together, and they’ll divide the benefit sum by the risk sum. And if the division is bigger than one, they’ll say the benefit’s bigger than the risk. And if the division is less than one, they’ll say the risk is bigger than the benefit.
Transforming Requirements Engineering with AI to Enhance Clarity, Consistency, and Scalability
As systems grow more complex, traditional processes struggle to keep up, ultimately impacting requirements quality. AI can assist in processing the sheer volume of data, enhancing clarity, consistency, and scalability across workflows.
Join Katie Huckett, Product Line Manager for Advisor/AI at Jama Software, for an exclusive webinar exploring how AI is becoming an essential cognitive amplifier in requirements engineering. Discover how AI is redefining the way teams detect ambiguity, surface hidden conflicts, and maintain alignment at scale.
What You’ll Learn:
Understand why requirements quality is declining under modern system complexity.
Learn the hidden costs of poor requirements and why traditional practices fall short.
Discover how AI amplifies cognitive processing and improves requirements quality.
Explore practical steps for adopting AI in your engineering workflows.
Gain insights into the future of requirements engineering with AI.
The video below is a preview of this webinar, click HERE to watch it in its entirety
WEBINAR TRANSCRIPT PREVIEW
The Collapse of Requirements Quality Under System Complexity – How AI Can Help
Katie Huckett: Welcome, and thanks for joining. Today we’re going to talk about something many engineering organizations are experiencing, but rarely say out loud. Requirements quality is collapsing under the weight of modern system complexity. This session isn’t about tools, features, or automation for automation’s sake. It’s about why this problem exists, why traditional fixes are no longer sufficient, and why AI is becoming a necessity rather than a nice to have in requirements of engineering.
My name is Katie, and I lead product strategy focused on AI-driven capabilities and requirements management. I spend most of my time working with engineering teams in highly regulated complex industries, aerospace and defense, automotive, medical devices, and other systems where requirements quality is not optional. What I’m sharing today is based on what those teams are actually struggling with in practice, not theory.
Here’s how we’ll spend our time together. We’ll start looking at why requirements quality is breaking down despite increased process maturity. We’ll talk about the hidden costs of complexity and why traditional approaches no longer scale. Then we’ll look at how AI changes what’s possible, not as a replacement for engineers, but as a cognitive amplifier. And finally, we’ll discuss what this shift means for engineering organizations moving forward. We’ll have a brief Q&A portion before we conclude today. Let’s dive in.
Here’s the paradox we’re living in. Requirements practices are more mature than they’ve ever been. Teams have invested heavily in process, tooling, standards, and governance, and yet many organizations are seeing more rework, more late stage surprises, and more friction between teams than before. What’s important here is that this isn’t happening because teams stopped caring about quality. It’s happening because the nature of the systems we’re building has changed faster than the way we manage requirements. In other words, the rules of the game changed, but most practices did not.
Modern products are no longer confined to a single domain. A single system now routinely spans software behavior, physical components, data flows, safety constraints, regulatory requirements, and operational considerations. All of these elements evolve together, often on different timelines and often with different teams responsible for each part. As systems scale and change in parallel, the number of relationships between requirements increases dramatically, not linearly. And yet, many traditional approaches still assume that these relationships can be reasoned through manually during periodic reviews or checkpoints. The challenge isn’t capability or commitment. It’s that the structure of the work itself has fundamentally changed.
Huckett: Before we go further, I want to ground this discussion in your experience. We’re going to launch a poll. Please take a moment to answer honestly. What is the biggest contributor to requirements quality issues in your organization?
Looks like we have the results in. In nearly every organization I work with, the answer is rarely just one of these. These challenges stack on top of each other, and that compounding effect is exactly what overwhelms traditional requirements practice.
Traditional requirements practices were built for a world where change was slower, and systems were more predictable. Reviews happened at defined milestones. Documents were relatively stable. Dependencies were fewer and easier to reason about. Today, however, requirements are changing continuously, often across teams working in parallel. When you apply periodic document-centric review models to this environment, gaps are almost inevitable. The process itself isn’t wrong. It’s just being asked to operate outside the conditions it was designed for.
It’s important to say this clearly. This is not a lack of skill problem. It’s not a lack of effort problem. It’s not a lack of accountability problem. It’s a structural mismatch between human cognitive limits and the complexity of modern systems.
One of the most dangerous things about requirements quality issues is that they rarely fail loudly. A single ambiguous requirement doesn’t stop a project. It quietly creates multiple interpretations. Those interpretations propagate into design decisions, test cases, and validation activities. By the time the issue is discovered, multiple teams have already invested time and effort based on different assumptions. And at that point, the cost isn’t just fixing the requirement. It’s undoing everything that was built on top of it.
Huckett: Let’s do another quick poll. Where do requirements quality issues most often surface too late in your lifecycle?
Some interesting results here. Wherever this shows up in your lifecycle, the pattern is consistent. Humans don’t see the issue until it’s already costly. That’s not a vigilance problem, that’s a visibility problem. When quality issues surface, the instinctive response is to add more safeguards. That means more reviews, more sign-offs, more documentation. The problem is that these measures increase effort without increasing visibility. Teams end up spending more time checking artifacts, but not necessarily improving quality or alignment. In highly complex systems, quality doesn’t improve by adding friction. It improves by improving signal.
This is where AI fundamentally changes the equation. AI doesn’t get tired. It doesn’t lose focus. It doesn’t skip over sections because a document is long or familiar. It can continuously scan requirements, compare them, and look for patterns or anomalies across the entire system. That doesn’t replace human expertise. It supports it by ensuring that engineers are spending their time where judgment actually matters. In that sense, AI becomes part of the engineering infrastructure rather than a separate tool.
2026 Predictions for Nuclear Energy: Innovation, Safety, and the Path to a Sustainable Future
The nuclear energy industry stands at a pivotal moment where innovation and tradition intersect to tackle the world’s most urgent challenges: decarbonization, energy security, and sustainability. From the emergence of small modular reactors (SMRs) and advanced reactor designs to the adoption of AI, automation, and digital engineering, the sector is embracing transformative technologies that are set to redefine how nuclear power is designed, operated, and perceived.
Key trends shaping the nuclear landscape include the transition from conceptual innovation to deployable solutions, the role of digitalization in enhancing safety and efficiency, and the evolution of regulatory frameworks to support next-generation technologies. Additionally, cybersecurity, workforce development, and global collaboration are becoming essential pillars of the industry’s future, ensuring that growth and innovation remain firmly grounded in the safety-first principles that define nuclear energy.
In this final blog of the 2026 prediction series, we bring these insights to life with perspectives from Jama Software’s industry expert, Patrick Garman, Solutions Manager for Energy, Industrial, and Consumer Electronics sectors. Patrick shares a forward-looking vision for 2026 and beyond, exploring the deployment of SMRs and advanced fuels, the integration of predictive analytics and real-time monitoring, and the innovations, strategies, and cultural shifts that will shape the nuclear industry’s role in a clean energy future.
Curious to read leading thought leaders’ predictions for their industries in 2026 and beyond? Dive into each blog below:
Q: What next-generation technologies (e.g., small modular reactors, advanced reactor designs, digital control systems) will have the most significant impact on the nuclear industry in the next five years? How can organizations prepare to adopt and regulate these innovations safely?
Patrick Garman: Over the next five years, the nuclear industry is likely to be shaped by a practical shift from conceptual innovation to deployable technology. Small modular reactors (SMRs) and microreactors are expected to lead this transition, moving beyond pilot projects toward early commercial use thanks to their modular construction, smaller footprints, and ability to serve diverse applications, from grid support to industrial process heat and remote operations. In parallel, advanced non-light-water reactors, such as high-temperature gas, molten salt, and fast reactors, are gaining traction as long-term solutions for high-efficiency power generation and emerging use cases like hydrogen production and industrial decarbonization. These reactor designs are closely linked to advanced fuels, including HALEU and TRISO, making fuel availability, qualification, and supply chain readiness a central factor in how quickly projects can move forward. At the same time, the industry is embracing digital instrumentation and control, automation, and data-driven operations to improve performance, reliability, and safety while also introducing new considerations around software assurance and cybersecurity. Underpinning all of this is a growing reliance on factory-based manufacturing, modularization, and robotic inspection, which promise to reduce construction risk and improve quality, provided these methods can be consistently qualified and aligned with regulatory expectations.
Safety and Risk Management
Q: Safety has always been central to the nuclear industry. How can digitalization, real-time monitoring, and predictive analytics further strengthen plant safety and reliability? What cultural or procedural shifts are needed to sustain a modern safety-first approach?
Garman: Digitalization is giving the nuclear industry new ways to reinforce its longstanding safety-first foundation by improving visibility, consistency, and foresight across plant operations. Real-time monitoring and predictive analytics allow operators to detect early signs of equipment degradation, performance drift, or abnormal conditions well before they escalate into safety or reliability concerns, while modern digital control and decision-support systems help reduce human-factor risk by delivering clearer, more contextual information during both normal and off-normal operations. To fully realize these benefits, organizations must evolve their safety culture and procedures to treat software, data, and analytics as safety-relevant assets governed with the same rigor as physical systems, while strengthening the human-automation partnership through training, validation, and clear operational boundaries. A modern safety-first approach, therefore, extends beyond traditional engineering excellence to include disciplined digital governance, cybersecurity resilience, and continuous learning, ensuring that advanced technologies enhance the conservative decision-making that defines nuclear safety.
Digital Modernization
Q: How do you see digital engineering and integrated data environments improving plant lifecycle management, from design through decommissioning? What challenges exist in migrating from legacy systems to modern digital platforms?
Garman:Digital engineering and integrated data environments are changing how nuclear plants are managed across their entire lifecycle, helping teams maintain clarity and control from early design decisions all the way through operations and eventual decommissioning. By creating a connected digital thread that links requirements, design models, safety analyses, construction records, and operational data, organizations can avoid the information loss that often happens at handoffs between phases or teams. This continuity makes it easier to manage design changes, maintain configuration control, respond to regulatory questions with confidence, and use operational insight to plan maintenance, life extensions, or decommissioning activities more effectively.
The biggest challenge is not the technology itself, but the transition. Many nuclear organizations are working with decades of legacy systems, documents, and institutional knowledge that were never designed to work together. Migrating to modern digital platforms requires careful, phased approaches that preserve trust in the data, maintain regulatory confidence, and respect the realities of long-lived assets that cannot pause operations for wholesale transformation. Success depends on strong data governance, disciplined change management, and a clear understanding that digital modernization is a long-term capability investment.
Regulatory and Compliance Evolution
Q: As global interest in nuclear energy grows, particularly across the EU, how can the industry ensure regulatory frameworks keep pace with innovation? What best practices can help organizations streamline compliance without compromising safety?
Garman: As interest in nuclear energy accelerates, the challenge is ensuring regulatory frameworks evolve alongside innovation without undermining the industry’s uncompromising safety standards. New reactor designs, fuels, and digital technologies don’t fit neatly into licensing models that were built around large, conventional plants, which means regulators and industry alike must continue shifting toward risk-informed, technology-inclusive approaches. This evolution works best when developers engage regulators early and often, clearly articulate their safety case, and align on expectations for evidence, review milestones, and decision points before designs are finalized.
Best-in-class organizations are streamlining compliance by treating it as an integrated engineering discipline rather than a late-stage documentation exercise. That means embedding regulatory requirements directly into design and development workflows, maintaining clear traceability from safety objectives to implementation and verification, and reusing proven arguments, data, and analyses wherever possible. At the same time, harmonization efforts across jurisdictions, transparent regulatory collaboration, and disciplined change control help reduce duplication without sacrificing diligence. The result is a more predictable path to licensing that supports innovation while preserving the conservative, safety-first principles that underpin public trust in nuclear energy.
Q: As nuclear facilities adopt more connected technologies, how can organizations guard against cyber threats while maintaining system integrity and safety? What proactive measures should become industry standard?
Garman: As nuclear facilities adopt more connected and digital technologies, cybersecurity is becoming inseparable from plant safety and reliability. Guarding against cyber threats starts with treating operational technology as safety-relevant infrastructure that is designed from the outset to limit the impact of any compromise through strong segmentation, controlled data flows, and isolation of critical functions. Leading organizations focus less on individual tools and more on disciplined system architecture, configuration control, and integrity protection, ensuring that digital systems behave predictably even under adverse conditions.
The industry is converging on practices such as secure-by-design engineering, rigorous access and change management, continuous monitoring tailored to OT environments, and well-rehearsed incident response that includes operations and engineering, not just IT. Ultimately, sustaining system integrity in a more connected nuclear plant depends on a cultural shift that recognizes cybersecurity as an extension of nuclear safety itself, governed with the same conservative mindset and operational rigor that public trust in the industry depends on.
AI and Automation
Q: What role will AI and automation play in improving design and manufacturing of nuclear reactors and efficiency, safety inspections, and predictive maintenance across nuclear facilities? What safeguards are needed to ensure responsible, transparent use?
Garman: AI and automation are set to play an increasingly practical role in the nuclear industry, particularly in areas where consistency, pattern recognition, and early detection matter most. In design and manufacturing, AI-assisted analysis can help engineers explore design alternatives, identify potential safety or manufacturability issues earlier, and improve quality through automated inspection, welding verification, and non-destructive evaluation. Across operating plants, automation and advanced analytics support more efficient inspections and predictive maintenance by detecting subtle equipment degradation, prioritizing risk-significant issues, and reducing unnecessary exposure of personnel to hazardous environments. Used appropriately, these technologies strengthen safety and reliability by helping teams act earlier and with better information.
Responsible AI deployment in the nuclear industry means applying the same conservative, evidence-based mindset that governs other safety-relevant systems. That means clearly defining where AI provides decision support versus where humans retain authority, validating models against real-world data, monitoring performance and drift over time, and maintaining full transparency into how recommendations are generated. Strong data governance, configuration control, and cybersecurity protections are essential, as is the ability to audit and explain outcomes to regulators and operators alike. When paired with clear safeguards and human oversight, AI and automation can become trusted tools that enhance the nuclear industry’s long-standing commitment to safety and public confidence.
Sustainability and Public Perception
Q: How can the nuclear industry strengthen public trust while positioning itself as a key player in the clean energy transition? What strategies are most effective for communicating safety, sustainability, and innovation to the public?
Garman: Strengthening public trust is ultimately about consistency between what the nuclear industry says, what it does, and what people experience over time. As nuclear positions itself as a critical enabler of a reliable, low-carbon energy system, the industry has an opportunity to connect its long-standing safety culture with today’s clean energy priorities, emphasizing not just carbon-free electricity, but resilience, energy security, and long-term environmental stewardship. Trust grows when organizations are transparent about both benefits and risks, communicate clearly how safety is engineered and governed, and demonstrate that lessons learned are actively shaping modern designs and operations.
The most effective communication strategies focus on clarity, credibility, and relevance to everyday concerns. That means moving beyond technical jargon to explain safety, waste management, and sustainability in plain language, using real data and independent validation rather than promises. Engaging early and continuously with communities, regulators, and policymakers helps demystify nuclear technology and humanize the people behind it. By pairing transparent communication with visible innovation, strong regulatory oversight, and measurable climate impact, the nuclear industry can reinforce public confidence while positioning itself as a trustworthy and essential contributor to the clean energy transition.
Q: With a generational shift in the workforce, how can the nuclear industry retain institutional knowledge while equipping new engineers with the digital and safety-focused skills needed for the next era of nuclear operations?
Garman: The challenge is to preserve institutional knowledge while preparing a new generation of engineers to operate in a far more digital, data-driven environment. Leading organizations are addressing this by deliberately capturing design intent, operating experience, and lessons learned in structured, accessible formats while pairing this with training that blends systems thinking, digital engineering tools, and a strong grounding in nuclear safety culture. Mentorship, cross-generational teams, and scenario-based training help bridge experience with innovation, reinforcing conservative decision-making even as new technologies are adopted. Ultimately, success depends on treating knowledge management and workforce development as long-term strategic investments, ensuring that the next generation has access not just to new tools, but have the mindset and discipline that have define safe nuclear operations.
Global Collaboration and Standardization
Q: How can international collaboration and harmonized safety standards support the safe expansion of nuclear energy, particularly as more nations revisit nuclear as part of their net-zero strategies?
Garman: Collaboration and harmonized safety standards are essential to the safe and timely expansion of nuclear energy as more countries turn to nuclear power to meet net-zero goals. Shared regulatory principles, common safety objectives, and mutual recognition of technical assessments help reduce duplication, improve consistency, and raise the global safety baseline – especially as new reactor technologies are deployed across multiple jurisdictions.
Collaboration among regulators, operators, and international bodies also accelerates the exchange of operating experience and lessons learned, allowing emerging nuclear programs to benefit from decades of global expertise. When paired with strong national oversight, this alignment supports innovation without compromising rigor, enabling countries to expand nuclear capacity with confidence while reinforcing public trust in nuclear safety worldwide.
Future Outlook
Q: What trends—technological, regulatory, or geopolitical—will most influence the global nuclear industry over the next decade? How can companies balance growth, innovation, and safety as nuclear energy plays a larger role in global sustainability goals?
Garman: Over the next decade, the global nuclear industry will be shaped by a convergence of technological innovation, evolving regulatory approaches, and shifting geopolitical priorities tied to energy security and decarbonization. Technologically, the progression of small modular and advanced reactors, digital engineering, and data-driven operations will expand where and how nuclear can be deployed, while fuel supply chains and cybersecurity will remain strategic constraints. Regulators are increasingly adapting frameworks to accommodate new technologies through risk-informed, technology-inclusive approaches, even as geopolitical dynamics, such as supply chain resilience, international collaboration, and regional energy independence, reshape investment and deployment decisions. To balance growth, innovation, and safety, companies will need to embed safety and compliance into their innovation processes from the outset, engage regulators and stakeholders early, and maintain disciplined governance over digital and organizational change. Those that succeed will be the ones that treat safety not as a brake on progress, but as the foundation that allows nuclear energy to scale credibly and sustainably in support of global climate and energy goals.
This achievement isn’t just a statistic; it is a direct reflection of the feedback, reviews, and success stories from our valued customers. In an industry where precision and traceability are paramount, being named the definitive leader ahead of competitors like Polarion, IBM® DOORS®, and Codebeamer validates our mission. We are here to modernize how complex products are built.
Our performance in the Winter 2026 report was exceptional. We secured a staggering 98% customer satisfaction score and expanded our recognition across multiple categories globally. This blog post explores what these accolades mean for our users and why Jama Connect continues to set the standard for modern requirements management.
Why G2 Rankings Matter
Before diving into the specific awards, it is important to understand the weight of a G2 report. Unlike traditional analyst reports that may rely heavily on theoretical positioning along with other factors, G2 is democratized. It is the voice of the actual user.
G2 rankings are based on two primary pillars:
Customer Satisfaction: This comes directly from verified user reviews, ratings, and feedback.
Market Presence: This accounts for market share, vendor size, and social impact.
To be named a Leader, a product must excel in both areas. It means the software is not only widely adopted but also deeply appreciated by the people who use it daily. When you see Jama Connect in the top right “Leader” quadrant, it signifies a tool that delivers on its promises, helping teams navigate the complexities of product development with confidence.
The headline of the Winter 2026 report is our Customer Satisfaction score of 98. In the software world, specifically in complex engineering sectors, users are often critical. They demand high performance, reliability, and ease of use. A near-perfect score is rare and hard-earned.
This score tells us that we are solving the right problems. It means that when engineering teams need live traceability, they find it intuitive. When compliance managers need to generate an audit trail, the process is seamless.
Beyond the Number
A 98% satisfaction score translates to tangible benefits for your team:
Reduced Friction: Teams spend less time fighting the tool and more time engineering.
Higher Adoption: When software is easy to use, team members actually use it, ensuring data integrity.
Trust: Users rely on Jama Connect as their single source of truth.
Leadership Across Every Business Segment
One of the most compelling aspects of the Winter 2026 report is the versatility of our leadership. Software often struggles to scale. Tools that are great for small startups often buckle under the weight of enterprise data, while enterprise tools can feel bloated and sluggish for smaller teams.
Jama Connect has proven it can handle it all. We earned Leader badges across three distinct business segments:
1. Enterprise Leader
Large organizations face unique hurdles, including massive data sets, rigorous security protocols, and thousands of concurrent users. Our leadership here confirms that Jama Connect scales effortlessly, supporting global enterprises without compromising on speed or performance.
2. Mid-Market Leader
Mid-sized companies are often in a critical growth phase. They need the rigor of enterprise compliance tools but the agility of a startup. We provide that balance, allowing mid-market teams to mature their processes without slowing down innovation.
3. Small-Business Leader
Even small teams tackle complex engineering challenges. Being a leader here shows that our platform remains accessible. You don’t need an army of administrators to get value out of Jama Connect; it is designed to be intuitive from day one.
Global Recognition: A Regional Leader
Innovation is global, and so is our user base. The Winter 2026 report highlighted our strengthening presence across international markets. We were named a Regional Leader in Asia Pacific, EMEA, and Europe.
This geographic diversity is crucial for modern product development. Many of our customers operate with distributed teams: engineering in Germany, manufacturing in China, and product management in the US. Our recognition as a global leader ensures that no matter where your team sits, Jama Connect provides a unified, high-performance experience.
Staying at #1 requires constant evolution. We are proud to be named a Momentum Leader, a badge that recognizes our continued growth and innovation trajectory.
Being in the top 25% of products for growth signals that we aren’t resting on our laurels. The landscape of systems engineering is changing rapidly with the introduction of AI, increasing regulatory pressures, and more complex hardware-software integration. Our momentum score proves that we are actively evolving the platform to meet these modern challenges head-on.
The Power of Relationships
While technical features are critical, the human element of software partnership is often overlooked. In the Winter 2026 report, Jama Connect earned the Best Relationship badge for Overall, Enterprise, and Mid-Market segments.
This recognition is perhaps the one we cherish most. It reflects the high-quality support, collaboration, and success services our customers experience. When you choose Jama Connect, you aren’t just buying a license; you are gaining a partner committed to your long-term success.
High relationship scores indicate:
Responsive Support: We are there when you need us.
Customer Success: We help you map your processes to the tool for maximum value.
Community Engagement: We listen to user feedback to drive our roadmap.
Voices from Our Community
The statistics are powerful, but the stories behind them are what truly matter. The praise and constructive feedback from our users on G2 drive our innovation. Here is what verified users are saying about their experience:
“Jama Connect is a powerful tool for requirements management and offers a wide range of features. Until now, the traceability of requirements was very difficult or even impossible. Jama Connect solves this problem 100%.”
— Verified User, Renewables & Environment, Enterprise
This review highlights a core struggle for many engineers: traceability. Without a tool like Jama Connect, tracing a requirement through to testing and validation is a manual, error-prone nightmare. We solve that.
“In the past, all requirement-related information was scattered across Jira, Confluence, Word, and Excel… With Jama Connect, we have centralized most of this information into a single, reliable source of truth.”
— Verified User in Manufacturing, Mid-Market
This user touches on the “document-centric” trap. Moving away from scattered Word docs and Excel sheets into a data-centric model is the single biggest leap a team can take toward maturity.
Driving Modern Requirements Management Forward
This G2 recognition reinforces our mission to help organizations move beyond outdated processes. The days of managing complex systems in static documents are over. Jama Connect provides a modern, Live Traceability™ platform designed for the complexities of today’s product development landscape.
With Jama Connect, teams can:
Improve Clarity: Establish a single source of truth that aligns all stakeholders.
Ensure Traceability: create live links across requirements, risks, and tests to visualize impact instantly.
Streamline Compliance: Automate the documentation needed for industry-specific regulations (ISO 26262, FDA 21 CFR Part 820, etc.).
Accelerate Cycles: Catch errors early in the process to avoid costly rework later.
Conclusion
We could not have achieved this milestone without you. Your partnership, feedback, and trust are the cornerstones of our success. These G2 awards, our eighth consecutive quarter at the top, are a shared victory for every team using Jama Connect to solve complex challenges.
Thank you for making us the #1 choice for requirements management. We are committed to continuing this journey of excellence with you.
Ready to see the full data? Download the G2 Winter 2026 Grid Report to explore why Jama Connect continues to lead the competition in Requirements Management.
Jama Connect® for Semiconductors Solution – It’s about time!
Jama Software recently announced the release of a solution tailored for the Semiconductor industry! This highly anticipated release has been eagerly awaited by both Jama Software’s Semiconductor team and customers, who are excited to see how it can address their requirements and data governance challenges.
As someone with a 24-year career in the semiconductor industry, primarily focused on Requirements Engineering for silicon and networking products, I’ve been anticipating this release for quite some time. My experience includes working with various tools, homegrown solutions, and the ubiquitous but limited Microsoft Word and Excel. Jama Connect stood out early on for its usability, ensuring engineering teams could adopt it without slowing down project cycles—a critical factor for success.
Addressing Semiconductor Complexity
The semiconductor industry’s complexity and variability have long posed challenges for requirements management tools. Fabless silicon design companies, EDA tool developers, and manufacturers all have unique data needs. A solution for this industry must be both configurable and governable. Unlike other industries, semiconductors lack a standardized information architecture, leading to locally optimized but unscalable solutions.
Jama Connect for Semiconductors addresses these challenges by offering a tailored, scalable solution. It provides a framework that balances flexibility with governance, making it easier for teams to manage requirements effectively.
In a recent video podcast (CLICK TO WATCH), Jama Software’s Semiconductor GM Neil Stroud described Jama Connect as the “incumbent standard” in the semiconductor industry. While the new solution was just launched, many semiconductor companies have already adopted Jama Connect as a step up from scattered document-based systems. Its SaaS cloud deployment, easy imports from Word or Excel, and support from Jama Connect Solutions Consultants enable teams to get up and running quickly.
Many companies have realized the need to modernize their requirements and data governance practices. Jama Connect’s usability, combined with expert consulting, has helped teams tailor the solution to their needs, even when starting with frameworks designed for other industries.
Time to Efficiency
A new Jama Connect deployment delivers value immediately. For example, a set of 100 requirements can be sent for review to 20 stakeholders in just a morning. The data architecture can be adjusted to match the current format, and documents can be imported into the tool. Reviewers, even those without Jama Connect licenses, can participate for free. This streamlined process reduces the time and effort required for reviews.
Jama Connect Advisor™ further enhances efficiency by identifying ambiguities and errors in requirements, improving clarity before reviews begin. This reduces the time reviewers spend asking clarifying questions, speeding up the entire process.
Deciding on a requirements management solution can be a lengthy process for large organizations. However, a Jama Connect for Semiconductors SaaS pilot can be configured in days, complete with SOC2 and other security features. Jama Software’s Solutions Consultants work with teams to define the scope of the pilot and establish success metrics, ensuring a focused and effective evaluation.
Time to Value
Semiconductor companies need to see tangible benefits from deploying Jama Connect. Executives and senior leaders value predictable cycle times and reduced costs from earlier defect identification. Over time, as more data is governed within Jama Connect, product definitions become clearer, and schedule volatility decreases. The tagline “Visibility leads to Accountability” highlights how Jama Connect provides leaders with progress indicators to monitor and address issues proactively.
Engineering managers benefit from a deployment path that doesn’t disrupt tight schedules. Jama Connect’s test capabilities, including a new feature in beta that generates test cases from requirements, help teams assess product quality and readiness. Integrations with common semiconductor tools and the Jama Connect Interchange API enable a comprehensive view of the development process.
For engineers, the intuitive user interface and immediate benefits of the Review Center help overcome initial skepticism. While transitioning from document-based systems to a modern requirements management solution requires effort, the long-term benefits are significant.
Time to Get Started
Looking back on my career in semiconductor requirements engineering, I often wished for a purpose-built solution like Jama Connect for Semiconductors. This solution offers the flexibility and governance needed to address the industry’s unique challenges. If you’re interested in learning more, Jama Software is ready to help. Current customers can reach out to their Customer Success Manager, while new prospects can contact their Account Executive to schedule a demo and discuss next steps.
2026 Predictions for AECO: AI, Digital Twins, and the Path to Sustainable Transformation
As we step into 2026, the Architecture, Engineering, Construction, and Operations (AECO) industry is poised for a transformative leap. From the integration of AI and digital twins to the adoption of robotics and advanced materials, the sector is embracing innovation to tackle its most pressing challenges: sustainability, efficiency, and collaboration in a hybrid world.
This year’s predictions explore how emerging technologies like generative design, predictive analytics, and automation are reshaping the project lifecycle. We’ll dive into the role of advanced digital tools in achieving net-zero goals, the growing importance of cybersecurity in a connected ecosystem, and the long-term trends that will define the industry for years to come.
In part six of this year’s predictions series, we bring these insights to life with perspectives from Jama Software’s own AECO experts: Joe Gould – Senior Account Executive, and Michelle Solis – Associate Solutions Architect, who share their vision for the future. From AI-driven decision-making to the rise of modular construction and lifecycle optimization, this piece highlights the innovations and strategies that will shape 2026 and beyond.
Curious to read leading thought leaders’ predictions for their industries in 2026 and beyond? Dive into each blog below and stay tuned for part 6, the finale of this year’s series:
What specific emerging technologies (e.g., AI, digital twins, generative design, robotics) do you believe will have the most transformative impact on the AECO industry in the next five years? How can firms prepare to adopt and integrate these technologies effectively?
Joe Gould: AI and Machine Learning will become foundational across the entire project lifecycle.
Design & Planning: AI accelerates generative design by evaluating thousands of options against constraints like cost, performance, and sustainability—helping teams reach optimized solutions faster.
Predictive Insights: By analyzing large datasets, AI can forecast risks, schedule impacts, cost overruns, and potential failures, enabling earlier and more informed decisions.
Workflow Automation: Routine tasks such as data entry, document review, and quantity takeoffs are increasingly automated, allowing teams to focus on higher-value, strategic work.
Digital Twins extend these capabilities into operations.
Operational Optimization: Real-time digital replicas of assets enable continuous monitoring and simulation, improving energy performance, asset utilization, and long-term operating costs.
Predictive Maintenance: Simulating asset behavior under different conditions helps identify issues before failure, reducing downtime and extending asset life.
Collaboration: A shared, real-time data environment ensures all stakeholders are aligned on the most current information throughout the asset lifecycle.
Robotics and Automation have been moving from experimentation to real jobsite adoption.
On-Site Execution: AI-enabled robotics handle repetitive and high-risk tasks with greater precision and safety.
Autonomous Equipment: Drones and self-operating machinery are increasingly used for surveying, inspections, and material movement, improving efficiency while reducing labor constraints.
Sustainability and Net-Zero Goals
With the AECO industry under increasing pressure to meet sustainability and net-zero targets, what role do you see advanced software, materials innovation, and digital tools playing in achieving these goals? Are there specific technologies or strategies you think will lead the way?
Gould: Important question! Advanced digital tools allow teams to understand and manage environmental impact early in the process, long before construction begins.
At the core is Building Information Modeling (BIM), which provides a data-rich model that supports ongoing analysis of energy performance, material use, and constructability as designs evolve. Energy modeling and simulation extend this by forecasting real-world performance early, allowing teams to optimize efficiency and integrate renewables before decisions are locked in.
AI and machine learning add another layer by analyzing large datasets to improve decision-making, optimize resources, and surface risks earlier. Generative design helps teams evaluate thousands of design options that balance sustainability, cost, and performance. Digital twins, fed by real-time sensor data, carry this forward into operations—enabling predictive maintenance, smarter energy management, and continuous performance optimization over the life of the asset.
Life-cycle assessment tools tie it all together by informing material choices based on embodied carbon and long-term environmental impact, not just upfront cost.
Materials innovation focuses on reducing embodied carbon and supporting a more circular approach to construction.
This includes a shift toward low-carbon materials such as mass timber, green steel, and advanced concrete alternatives, along with greater use of recycled and reusable content. High-performance insulation and composites further improve operational efficiency by reducing long-term energy demand while maintaining durability and performance.
The real impact comes from integrating these tools into a single, data-driven approach—connecting design, construction, and operations.
Key strategies:
Data-driven decarbonization, using reliable project data for transparent reporting and continuous optimization
Prefabrication and modular construction, reducing waste, emissions, and schedule risk
Circular design principles, enabling reuse and recovery at end of life
Predictive maintenance, extending asset life and reducing long-term operational waste
By aligning digital tools, materials innovation, and lifecycle thinking, the industry can move beyond incremental gains and make measurable progress toward net-zero and long-term sustainability goals.
As hybrid and remote work models continue to evolve, how do you see these changes impacting collaboration, innovation, and project delivery in the AECO industry? What tools or processes will be critical for maintaining efficiency and creativity?
Gould: Hybrid and remote work are reshaping AECO, driving efficiency, expanding access to talent, and accelerating digital adoption—but they require more discipline around how teams collaborate and deliver work.
Collaboration has shifted from informal to intentional. Cloud-based platforms, shared models, and virtual design reviews are now standard, enabling distributed teams to stay aligned without being co-located. Innovation hasn’t slowed—it’s evolved. Access to broader talent pools and increased automation of routine tasks allow teams to spend more time on higher-value problem-solving.
From a delivery standpoint, hybrid models often reduce cycle times and costs. Work continues across time zones, travel is minimized, and documentation improves because communication has to be clearer by default.
Success in this environment depends less on tools alone and more on how they’re used. Cloud BIM, collaboration platforms, and project management systems form the backbone, but clear communication norms, standardized workflows, and outcome-based accountability are what keep teams productive.
To me, the shift isn’t about where people work—it’s about building repeatable, digital-first processes that support speed, clarity, and consistent project outcomes.
AI and Automation
How do you foresee AI and machine learning shaping decision-making, risk management, and project optimization in AECO? What are the biggest challenges or limitations the industry might face in scaling these technologies to automate processes?
Michelle Solis: While AI itself will make an impact on AECO companies, one additional area where we will see impact is in building the infostructure to handle the increase of AI usage across all industries. This will mean more jobs, job sites, data centers, and projects.
Gould: AI and machine learning are shifting AECO from reactive to proactive. When applied well, they improve decision-making, surface risk earlier, and optimize how projects are planned, built, and operated.
AI helps teams make better decisions by analyzing large volumes of historical and real-time data—highlighting patterns and risks humans typically miss. Generative design accelerates this by evaluating thousands of options against constraints like cost, performance, and sustainability. On the risk side, predictive analytics and real-time monitoring help identify schedule, cost, and safety issues before they escalate. AI also drives operational gains through task automation, smarter maintenance planning, and more resilient supply chains.
The challenge isn’t the technology—it’s scaling it. Most AECO firms struggle with fragmented data, limited system integration, and inconsistent standards. There are also a real skills gap and natural resistance to changing long-standing workflows. Add in high upfront costs, unclear use cases, unclear ROI, and legitimate concerns around data privacy and accountability, and adoption slows quickly.
The opportunity is real, but success depends on getting the fundamentals right: clean data, integrated systems, clear ownership, and practical use cases that tie directly to project and business outcome
Responsible AI Adoption
As AI and machine learning become more integrated into AECO workflows, what challenges or considerations should companies be mindful of to ensure successful implementation? How can firms address these challenges while maximizing the benefits of these technologies?
Gould: AI adoption in AECO isn’t a technology problem—it’s a fundamentals problem. Success depends on data, people, and how firms manage change.
Most organizations struggle with fragmented data, legacy systems, and limited AI-ready skills. Add natural resistance to new workflows, unclear ROI, and concerns around data security and accountability, and progress stalls quickly.
The path forward is straightforward:
Get the data right: standardize, govern it, and make it accessible
Upskill teams: treat AI as a productivity multiplier, not a replacement
Start small: focus on high-impact pilots that prove value fast
Modernize platforms: move toward cloud-based, integrated systems
Keep humans in the loop: clear ownership, transparency, and oversight matter
Firms that focus on these basics will scale AI effectively—and turn experimentation into measurable business outcomes.
Data-Driven Project Management
With the growing emphasis on predictive analytics, real-time monitoring, and data-driven decision-making, what strategies would you recommend for AECO firms to better harness data for optimizing project outcomes and resource allocation?
Gould: To use data effectively, AECO firms need to focus less on dashboards and more on fundamentals: integrated systems, clean data, and teams that actually trust and use it.
That starts with moving off siloed tools and spreadsheets and into cloud-based, integrated platforms that create a single source of truth across design, delivery, and operations. Strong data governance—clear ownership, standards, and quality controls—is non-negotiable. Without clean, consistent data, analytics don’t matter.
From there, predictive analytics should be embedded directly into project workflows, not buried in reports. Tracking the right KPIs and using data to flag schedule, cost, safety, and resource risks early shifts teams from reactive to proactive.
Finally, this only works if people are brought along. Start small with high-impact use cases, involve field teams early, and invest in basic data literacy, so insights drive decisions—not just meetings.
What upcoming regulatory changes or compliance requirements do you anticipate having the biggest impact on the AECO industry in 2026? How can companies stay ahead of these changes?
Gould: The biggest regulatory shifts hitting AECO in 2026 will center on ESG (Environmental, Social, and Governance), energy performance, and digital risk. ESG reporting is moving from “nice to have” to mandatory, with climate disclosure requirements cascading through supply chains. Energy codes will continue tightening, pushing firms toward higher-performance, low-carbon, and “zero-ready” buildings. At the same time, increased use of AI and cloud platforms is driving new expectations around transparency, governance, and cybersecurity.
The firms that stay ahead won’t treat this as a compliance exercise. They’ll lean on digital platforms to track energy, carbon, and materials from design through operations, put clear AI and data governance in place, and strengthen cybersecurity practices as reporting requirements tighten. Just as important, they’ll build regulatory awareness into project planning early—before requirements show up as cost, schedule, or risk surprises.
Cybersecurity in AECO
As digital tools and connected systems become more prevalent in AECO, what role do you see cybersecurity playing in protecting sensitive project data and ensuring operational continuity? Are there specific threats or solutions companies should prioritize?
Solis: As digital tools, connected platforms, and AI become more embedded in AECO workflows, cybersecurity will play a critical role in protecting sensitive project data and maintaining operational continuity. With the growing use of AI, firms must clearly define what data can and cannot be shared with AI models, particularly when working with proprietary designs, client information, or critical infrastructure data.
Beyond data leakage, organizations also need to address risks such as AI hallucinations, bias, and model misuse, which can directly impact design decisions, safety, and compliance if left unchecked. To mitigate these risks, companies should prioritize strong access controls, data governance policies, employee training, and secure AI deployments. Establishing clear guidelines around AI use, along with continuous monitoring and validation of outputs, will be essential to ensuring both cybersecurity and trust in digital systems as adoption accelerates.
Future of Innovation
What is the most innovative trend, tool, or process you’ve seen in the AECO industry recently? How do you anticipate it influencing the industry in the coming years?
Solis: One of the most impactful trends I’ve seen recently is the increased focus on Requirements Management across rail and broader AECO organizations. While this shift is often driven by hard lessons such as losing a contract or discovering unmet requirements late in a project, it signals a growing recognition that informal or disconnected requirement processes are no longer sustainable for complex, regulated projects.
Gould: The most meaningful innovation in AECO is the convergence of AI, digital twins, and integrated platforms. Together, they’re turning projects into connected, data-driven systems that move teams from static modeling to prediction, automation, and lifecycle optimization.
At the center is the digital thread. Requirements are no longer buried in PDFs and spreadsheets—they’re connected directly to BIM, schedules, costs, and real-time performance data. AI continuously validates designs against requirements, flags deviations early, and maintains traceability from concept through operations. That shift alone reduces rework, misalignment, and late-stage surprises.
AI-powered digital twins then extend this into delivery and operations, keeping stakeholders aligned and enabling smarter, faster decisions. The result is leaner execution, better compliance, and assets that actually perform as intended—not just on day one, but over their full lifecycle.
Long-Term Trends
What trends or technologies do you think will still be shaping the AECO industry five years from now? Ten years? How can companies position themselves to remain competitive in the long term?
Solis: I don’t think there’s one technology specifically that will shape the AECO industry. Companies who make an effort to welcome new technologies and not go against them will see success. This industry doesn’t want to evolve, but it will.
Gould: Over the next 5–10 years, AECO will be defined by digital maturity and industrialization. AI, BIM, and digital twins will move from tools to core infrastructure, while sustainability and offsite construction become standard, not optional.
In the next five years, BIM becomes the project command center—fully cloud-based and connected to schedule, cost, and lifecycle data. AI is embedded in planning and design to surface risk early, optimize decisions, and improve predictability. Modular and offsite construction scale quickly as firms respond to labor constraints and schedule pressure. Sustainability shifts from “nice-to-have” to a requirement.
Hard to say but looking ten years out I would predict that digital twins manage assets end-to-end, robotics handle more field execution, and buildings operate as connected systems within smart cities. Design, construction, and operations blur into a continuous, data-driven lifecycle.
The firms that win will invest early in integrated platforms, clean data, and workforce upskilling. They’ll focus on collaboration, specialization, and strong technology partnerships—turning digital capability into real project outcomes, not just innovation theater.
Engineering for the Cyber Resilience Act: Navigating Compliance Across the Product Lifecycle
Preparing for the Cyber Resilience Act: What Engineering Teams Need to Know Now
The EU Cyber Resilience Act (CRA) is setting new expectations for digital product development. It introduces mandatory requirements for vulnerability management, secure-by-design engineering, traceability, and post-market monitoring. For manufacturers of connected or software-enabled products, this represents a critical shift in how you build, document, and maintain your technology.
In this webinar,Patrick Garman, Manager of Solutions & Consulting at Jama Software, breaks down the complexities of the CRA, reviews enforcement timelines, and demonstrates how to integrate cybersecurity directly into your product lifecycle.
What You’ll Learn:
Deconstruct CRA Requirements: Gain a clear understanding of obligations for manufacturers, importers, and distributors, including secure development practices and vulnerability handling.
Operationalize Secure-by-Design: Learn practical strategies to embed security into your engineering workflows from day one.
Master Software Bill of Materials (SBOM) Transparency & Traceability: Discover how to maintain the rigorous documentation and traceability of the new regulation demands.
Navigate the Enforcement Timeline: Get a clear view of upcoming deadlines to help you prepare your organization strategically.
Leverage Jama Connect® for Compliance: Explore how a modern requirements management tool helps track threats, link mitigations to requirements, integrate testing, and prove compliance.
Don’t wait until the deadline approaches to address these critical changes. Watch now to ensure your team has the knowledge and tools to navigate the CRA successfully.
The video above is a preview of this webinar – Click HERE to watch it in its entirety!
TRANSCRIPT PREVIEW
Patrick Garman: Hi, everyone, and thank you for joining today. My name’s Patrick Garman, and I am the Solutions Manager for Energy, Industrial, and Consumer Electronics sectors here at Jama Software. Today, I’m going to be talking about the EU’s Cyber Resilience Act, or the CRA. I’ll explain what the CRA actually is, what it means for product developers, and how you can show evidence of secure by design without creating unnecessary overhead. I’m also going to briefly show how Jama Connect supports your CRA compliance. At a high level, the Cyber Resilience Act is an EU regulation that applies to products with digital elements, so hardware with software, firmware, or connectivity, and standalone software products as well. It’s not a technical standard, and it does not tell you how to implement security; it focuses on outcomes. Did you consider cybersecurity risks? Did you define mitigations? Can you show how those were implemented and maintained? It’s also worth saying what it’s not. It’s not saying that products must be perfectly secure, and it’s not trying to turn product teams into security researchers. It’s really about making cybersecurity part of normal product engineering, just integrating it into your process.
And the motivation behind the CRA is pretty straightforward: products today rely heavily on software, but cybersecurity practices across manufacturers vary a lot. Some teams are very disciplined, and others rely more on informal knowledge and experience. From a regulatory point of view, that makes it hard to assess product risk and hard to respond when vulnerabilities show up later, so the CRA is really about creating a consistent baseline, so cybersecurity is treated more like safety, reliability, or quality, something you design for, document, and revisit throughout the product lifecycle. And the penalties can be pretty stiff for non-compliance. You hear, for non-compliance, up to 15 million euros or 2.5% of your global annual turnover. Products can be barred from the EU market for non-compliance. It does include mandatory incident reporting, and it also establishes liability for manufacturers for unsafe or insecure products, so it is something that is very important to prepare for and be ready for. If you strip away the legal language, the CRA requirements really fall into a few practical buckets. First, you’re expected to identify cybersecurity risks that are relevant to your product and how it’s used.
Garman: Second, those risks should lead to actual security requirements, design constraints, controls, or behaviors that mitigate the risks. Third, there needs to be evidence, not just that you thought about security, but that the requirements were implemented and verified. And finally, the CRA expects manufacturers to manage vulnerabilities after release, things like intake, assessment, updates, and communication. And the challenge is doing it consistently and in a way that you can explain later, especially if this information is spread across different repositories. Before I jump into a demo in Jama Connect, I want to set up how to think about CRA compliance in Jama Connect. The CRA is ultimately asking for something pretty specific, can you prove a clean line from the cybersecurity risk to mitigation to verification, and then keep that story intact as the product changes? And Jama Connect’s a great tool for this because it’s designed for exactly this kind of lifecycle traceability with definable traceability information models that provide guardrails for your process. And the model I’m showing here, threats must link to one or more security requirements, and security requirements must link to verification evidence like test cases or analysis.
And if we want to go deeper, we can link into design and implementation artifacts as well. And the reason that this matters is that once these rules are in place, you’re not relying on memory or tribal knowledge. Jama Connect can guide teams towards consistent linking, and it becomes much easier to answer the questions that come up in audits and reviews, such as which risks are unmitigated, which mitigations aren’t verified, and what changed since the last release? And the other big benefit is the change impact. Sorry. When a new vulnerability pops up or a design decision shifts, Jama makes it practical to see what requirements, tests, and releases are affected without manually stitching it together across documents and spreadsheets. With that framing, what I’ll show next is a simple example. We’ll take a threat and author a requirement against it, and then see the verification evidence, so you’ll see how the relationship rule set keeps the traceability clean and reviewable. For this dem,o I’m going to keep the model intentionally simple. We’re going to start with a cybersecurity threat analysis, trace that to a security requirement, and then to a validation.
Garman: And in this scenario, I’m going to use the CVSS, which stands for the Common Vulnerability Scoring System, the 3.1 model, to score severity consistently. CVSS is traditionally used for vulnerabilities, but teams often use that same scoring structure for threat scenarios because it is familiar and repeatable. And I have a pre-created threat analysis item so that we can focus on the traceability aspects. But here you can see I have a place where I can provide a name, a description of the threat or vulnerability, and also select all of the appropriate vectors within the CVSS scoring model. And I’m also using Jama Connect Interchange™‘s Excel functions to calculate the base score and assign a severity rating, along with the temporal score and environmental score. Again, these are all calculated automatically on the backend as you define your threat vectors. And the reason I like capturing all of these attributes here in Jama Connect is it makes the assumptions explicit. Stakeholders can review the score, disagree with it, and adjust it, but we’re not hand-waving severity. And because it’s all on the same system as our requirements and validations, the cybersecurity story stays connected.
2026 Predictions for Semiconductors: AI, Chiplets, and the Path to Sustainable Innovation
As we step into 2026, the semiconductor industry stands at the crossroads of unprecedented technological advancements and complex global challenges. From the rise of AI-driven chip design and heterogeneous integration to the growing emphasis on sustainability and geopolitical shifts, the sector is navigating a transformative era.
The next wave of innovation will be defined by breakthroughs in advanced lithography, chiplet architectures, and quantum computing, while sustainability efforts will reshape manufacturing processes to address energy efficiency, water usage, and materials recycling. At the same time, the industry faces critical hurdles, including talent shortages, supply chain realignments, and the need for robust cybersecurity measures.
In this year’s predictions series, we’ve gathered insights from leading semiconductor experts:
Together, they explore the trends and technologies shaping the future of semiconductors. From AI-driven automation and edge computing to the challenges of regulatory shifts and the promise of chiplet-based architectures, this piece highlights the innovations and strategies that will define 2026 and beyond.
Q: What emerging technologies (e.g., advanced lithography, AI-driven chip design, quantum computing, heterogeneous integration) will have the most transformative impact on the semiconductor industry in the next five years?
Simon Bennett: In the next five years, the semiconductor industry will continue to grow, almost doubling in size from today to $1Trillion by 2030. But to sustain that growth, the industry will go through some extreme changes and challenges. The first trend to note is actually due to a declining trend as Moore’s Law continues to slow. [Editor’s note: Moore’s law is the observation that the number of transistors in an integrated circuit (IC) doubles about every two years.]
Moore’s Law has driven the growth of the Semiconductor industry for many decades, but it is bumping up against the fundamental laws of physics. The economics of scaling to the next node are increasingly prohibitive and taking longer and longer to reach fruition.
Whilst keeping an eye on what is coming out of China, there will be some more mundane but equally challenging technology trends that are emerging and will become increasingly important in 2026 and beyond. These are AI driven design, and both chiplet and wafer scale designs (two opposite ends of the spectrum, but both an engineering reaction to the slowing of Moore’s Law).
Neil Stroud: Given the ever-increasing innovation around AI and its associated deployment, chip development is under continued pressure to keep up. This is applicable across all architectures, including Central Processing Units (CPUs), Graphics Processing Units (GPUs), and Neural Processing Units (NPUs). Naturally, continued optimization will happen around acceleration and emerging technologies like process node shrinks (advanced lithography), AI-driven chip design, and the chiplet approach (heterogeneous integration). Process node shrinks will contribute. However, the chiplet approach will also drive heterogeneity across architectures and nodes. All these factors will intimately impact the next generation of chip families for AI in the datacenter and at the edge.
2: Sustainability and Manufacturing Efficiency
Q: How do you see sustainability influencing semiconductor manufacturing, particularly in areas like energy efficiency, water usage, and materials recycling? What strategies will help the industry achieve greener fabrication processes?
Bennett: This is a great question, and right now, the elephant in the room. From Fabs to datacenters, the environmental impact is huge. Water consumption alone is a huge factor. Twenty years ago, visionary realtors quietly purchased acres of land close to a bountiful supply of water and close to a large data pipe. Those realtors are now wealthy, and the secret is out. Now the price of that land is at a premium. So, the investors behind the fabs and the datacenters are using government subsidies and their own funds to find alternative sources of energy and resources. Nuclear is making a comeback, driven in part by the energy demands of the datacenters. Municipal areas like Phoenix are making guarantees of plentiful water to companies to attract them to their region; that will put them in direct conflict with farmers in California.
Most of this is happening off the radar of the mainstream media, and the political arena is presented as a battle for the best jobs. The concern over the environmental impact is not yet front and center. Two events will likely happen to change this:
The AI bubble will inevitably burst. Just like in the early days of the internet, there will be market correction as reality catches up to expectation. Just like the internet bubble, this doesn’t mean that AI is not going to be a societal change; it just means the market got too overheated.
Unfortunately, there will be some kind of accident related to the overbuild of the infrastructure around Datacenters and Fabs. A dam will burst (Phoenix – see Roosevelt Dam), or a multibillion fab will be damaged by a natural disaster (see fault lines in Taiwan). These two events will raise awareness of environmental costs relating to sustainability and manufacturing efficiency.
In other words, in the next five years, we will be forced to take a pause, a breath, and truly measure the value vs the cost. This isn’t a bad thing. Our human history of technology transformations is punctuated with these pauses and resets. Usually for the better.
Steve Rush: Sustainability is hugely influential and important. Energy demand is forecasted to accelerate with new data centers and the demand for AI. Semiconductor companies need a system to help manage their sustainability requirements and, very importantly, validate them. Implementation to hit targets and balance, power, efficiency, and sustainability will be a series of trade-offs – semiconductor organizations will need a tool to trace all of this information and prove that they meet sustainability targets and goals.
Sarah Crary Gregory: While the semiconductor industry is obviously fiercely competitive, it can match that intensity with fierce collaboration on critical issues. Sustainability is probably the most prominent area where industry consortia such as the Semiconductor Climate Consortium bring companies together to tackle common problems. Initiatives to enable water reclamation, reduce emissions, and produce data quantifying the return on investment of sustainability practices will be more critical with the burden placed on these resources from the exponential expansion of AI. The semiconductor industry is highly interdependent, and nobody believes that there’s a way to get a competitive advantage by monopolizing natural resources. The way forward is through innovations that decrease resource consumption and minimize waste, and initiatives for water reclamation/”net zero” resource use will continue to be essential investments.
Stroud: I think there are two parts to this. Firstly, the environmental impact of actually building the chips in foundries. A huge amount of effort and investment has gone into sustainability in semiconductor manufacturing, including energy efficiency, water usage, and materials recycling. semiconductor manufacturing and materials. A great example of this is massive recycling of water used in fab processes, as well as optimizing processes and the associated chemicals used, including minimizing atmospheric emissions.
Secondly, there is the environmental impact related to the deployment of the device itself, as it consumes power and emits heat. Of course, the extreme example of this is the data center where huge racks of GPUs or CPUs are deployed, collectively consuming Megawatts of power to both power them and cool them. Again, huge investment is going into driving data center efficiency. One way to contribute is through chip design optimization to improve ‘performance per Watt.’ That is simply a measure of how much computing can be done for a given Watt of power. This optimization can happen through design and architecture efficiencies as well as process node shrinks. Ensuring the software stack is also developed to drive efficient use of the underlying hardware platform also has a fundamental role to play. It’s easy to see that these steps can have a profound positive impact on the environment caused by the global electronics footprint.
Q: How is AI accelerating innovation in semiconductor design, verification, testing, and manufacturing? What challenges must companies overcome to fully leverage AI-driven automation?
Bennett: Natural language and agentic AI will continue to show up across the tool chain. But expect some resistance from SOC design engineers, who, ironically, since they are at the epicenter of the AI revolution, are traditionally conservative and slow to adopt new methods. Verification is the most in need of help with AI-driven automation, since there just aren’t enough engineers on the planet to drive the verification needs of an SOC. (see salaries on Glassdoor). It’s been estimated that with the use of AI, a team of 3 expert verification engineers can do the work of 5 traditional verification engineers with limited use of AI, in 3 to 5x less time. This is a compelling message to an (open-minded – see below for a caveat) engineering VP struggling to find the resources to deliver a fully validated product on time. These engineers and the tools they use will be in high demand in the next five years.
Beyond design, AI will show up in yield and manufacturing analytics. The challenge of inventory and yield management in the era of disaggregated chiplet-based designs is magnified. It’s essential that all the chiplets deliver the yield and volume needed at the exact same time. The overall package is only as good as the weakest tile. This is an underserved opportunity within the big three EDA companies, and the packaging OEMS tend to jealously protect their homegrown investments in solving these challenges. Expect emerging startups to come forward as disruptors in this particular segment in 2026 and beyond.
Rush: Every company is looking for ways to utilize AI in their organization. AI can play an important role in managing traceability, especially from siloed systems that are isolated from one another. Agentic experiences that improve engineer productivity really are key. The main challenge that AI has in the semiconductor space, in particular, is adoption with the engineering team. AI experiences must improve engineering productivity; they must be accurate, and they cannot be an impediment to use. If AI-generated content is of questionable quality or if the AI experiences become too burdensome to use, AI initiatives risk dying on the vine.
4: Supply Chain and Geopolitical Shifts
Q: How are global supply chain realignments and geopolitical factors shaping semiconductor strategy? What can companies do to mitigate risk and ensure resilience in developing complex products on their own or with co-development partners?
Bennett: A global supply chain developed over the past thirty years has delivered $1T in cost savings. This $1T is now under serious threat as the world is a very different place compared to when this globally interconnected environment was first conceived. In the next five years, expect China to become more self-sufficient as it replicates every aspect of what it previously relied on from overseas, from EDA to IP to fab equipment. Expect to see semiconductor-based products from coffee machines to phones to servers to (even) EVs sourced almost exclusively from China with little to no reliance on anything beyond the shores of China. This will trigger protectionist measures in the US and the EU as they work to protect homegrown industries from what will become increasingly consumer appealing products from the Chinese factories.
A more optimistic view may be that the tensions ease as the US / EU recognize the need for open trade with China, and continue to see its designs realized in Chinese factories (but I’m not holding my breath). In semiconductors, companies will be most susceptible to this shift in China as they move to homegrown alternatives. As the geopolitics ramp up, the focus on Provenance in the West will become a C-suite / US Senate / EU Parliament level of attention. Knowing where every component or piece of code originates, its genealogy will become paramount. A counterforce will emerge where the information is “buried” as the realization hits that we can’t possibly trace the root of every bit of code, every nanometer of design. Companies will emerge with one of two unique value propositions: 1) we can audit your product and provide the provenance, 2) everything you use is contaminated; we are a new company, built cleanly from the ground up. Somehow, all three will survive – the traditional companies, the auditors, and the new “clean” companies. But there will be some very interesting mergers and acquisitions, mostly off the radar as these three entities re-align and learn to co-exist.
Rush: These days, you can basically count on major geopolitical news covering the semiconductor industry week in, week out. At the end of the day, co-development and partnerships are key. The semiconductor supply chain is mind-bogglingly complex. Adopting modern, more collaborative tooling is on the rise. Historically, the semiconductor industry has even been hesitant to adopt cloud-based solutions, and I’ve definitely seen a change in the last few years around this.
Stroud: Like many other segments, the semiconductor market tends to be cyclic, which leads to times of undersupply and oversupply. This is a complex problem to manage with many factors, including global supply chain realignments and geopolitical factors. Naturally, foundry capacity has a big role to play, and we seem to be in an investment phase right now with a number of fabs being built. This is a massive investment with a modern fab costing tens of billions of Dollars and taking multiple years from construction start to mass production. Communication and collaboration across the ecosystem also has a role to play, especially now that we are accelerating into the chiplet era, which can help mitigate risk and ensure resilience in developing complex products.
5: Chiplet and Heterogeneous Integration
Q: What role will chiplet architectures and heterogeneous integration play in addressing performance and scalability challenges? What technical and ecosystem hurdles must be overcome?
Bennett: Chiplets are essential to the continued growth of Semiconductors. Without chiplets, the forecast CAGR ($1T by 2030) is unreachable (basic economics of Moore’s Law). The challenges are two-fold: 1) engineering challenges around interconnecting tiles from different suppliers running at high speed and with the thermal challenges of a modern chip; and 2) coherence – the coherence of the supply chain, compliance, and verification. More specifically, the standards emerging need to be better governed (e.g., Universal Chiplet Interconnect Express (UCIe) for interconnect and system architectures if they aren’t going to become bottlenecks stymying growth.
6: Talent and Workforce Development
Q: With growing global demand for skilled engineers and manufacturing specialists, how can companies address the talent shortage in the semiconductor industry?
Bennett: This is where AI needs to step in and become more readily accepted within Semiconductor Engineering orgs. As stated above, studies show that a small team of AI proficient verification engineers are 5x+ more productive than a traditional team. However, the resistance comes from within – engineers are conservative, and within a traditional engineering organization, the manager / Director / VP still measure their worth by the number of engineers the corporation is willing to fund. This leads to destructive behaviors, such as a VP of Verification Engineering employing 100 RTL validation engineers to do the job that 10 Functional Verification engineers could do because “it’s too expensive to hire the functional verification engineers” – the companies that will thrive and succeed in the next five years are the ones who break down this cultural impasse.
Rush: There are a lot of talented people in the job market right now who can help fill the gap. Hopefully, semiconductor companies will look to hire talent from across industries – automotive, medical, and aerospace. There are certain challenges in getting enough skilled foreign workers to fill certain roles – I’m more concerned that there are many highly skilled, talented people out there looking for jobs!
7: Regulatory and Export Controls
Q: How do evolving export controls, trade policies, and security regulations impact semiconductor innovation and competitiveness? How can companies adapt strategically?
Bennett: They don’t impact semiconductor engineering innovation or competitiveness – in fact, they improve it. Case in point is China – as access to advanced GPUS / EDA tools was limited, they innovated, and actually improved on the technologies they didn’t have access to. Another example is where the Russian engineers working for US companies prior to the war in Ukraine were let go and went to work for Russian companies, helping boost the AI business in Russia. But where the question applies is the innovation at the corporate level. Engineering innovation can be stymied by a C-suite overly concerned about trade or political issues. The paradox is that smaller companies with less of a global or political reach could feel less compelled to avoid the risk associated with innovation.
Gregory: “Evolving” is an understatement! The volatility around export controls and trade policy in the United States right now is simply unprecedented, and 2026 looks like more of the same. Companies can strategically navigate these unsettled times by implementing systems –people, processes, and tools – that enable maximum response flexibility. Modular architectures, whether they’re chiplet-based, specific configurations of IP cores, highly modular software, or other building blocks, will enable the development and delivery of products whose configurations can be changed and modified as circumstances warrant. Variant management is a critical capability to be able to swap features in and out based on policy changes. Solid, well-governed data foundations will be critical to stay on top of the wildly shifting policy landscape.
Q: As demand for edge AI and high-performance computing grows, what innovations are most critical to meet performance and power efficiency goals?
Bennett: There are many ways to answer this, but I’ll focus on the chip-level design aspect. First, the interconnect, as previously described – the clean adoption of UCIe and a strong governing body to oversee its evolution (think Universal Serial Bus, or USB.) 3D packaging needs to keep up with the thermal demands of a heterogenous package – this may lead back to the engineering talent pipeline previously discussed since the engineers who have the combination of skills to analyze and design (future-proof) these packages are unique (think warping of a substrate as it reacts to thermal pressures, leading to subtle issues with the interconnect manifesting as signal integrity.)
Rush: I’ll answer this more from a – data isolation – perspective. Design and testing are really important, but more important is tracing all the way to the highest level and validation. I think responsible AI will help with efficiency here, but companies need a way to trace from the top down. In all honesty, this is a challenge for the semiconductor industry – having one single source of truth that can prove you’re hitting sustainability goals.
9: Cybersecurity and IP Protection
Q: With increasingly complex global supply chains, how can semiconductor companies protect intellectual property and secure their design-to-production ecosystems?
Bennett: Expect a lot more reference to initiatives such as Software Bill of Materials (SBOM) and Engineering Bill of Materials (EBOM.) Expect the concept of a Bill of Materials (BOM) to evolve and take on more significance in the next few years. Expect the term Provenance to take on more importance. Traditional PLM companies will position themselves as the answer, but there will be significant pushback from the semiconductor industry, and rightly so – these PLM systems were never developed with semiconductors in mind. They are monolithic in nature, expecting the end user to move their data into their environments. The C-Suite will sign on, the engineers won’t. This will lead to QMS and IT organizations emerging to manually clone the data inside the PLM systems. For a while, this will seem just fine, until one or more issues come to public light, and the C-suite exec realizes they have spent a lot of money on tools and resources, and it didn’t solve the problem. Those companies that invested in a more lightweight engineer-friendly solution, providing traceability, compliance, and coherence insights without the costly overhead of monolithic tools and the resources that go along with them, will grab the attention of those who lost out. And yes, AI will play a part. A well-managed digital thread with the ability to expose itself in a controlled manner to intelligent insights will win out.
Rush: I mentioned earlier that semiconductor companies are adopting more cloud-based tooling. But they are not slacking in terms of security needs. By selecting best-in-class tools with exceptional infosec track records (like Jama Connect), they are effectively balancing speed and agility with security and not sacrificing either. They are pushing their vendors to expand their tool sets to deliver best-in-class experiences with rationale, scalable permission structures that are tightly governed. They’re looking for tools and vendors that are putting AI at the center of their vision – but need their vendors to offer closed, secure LLMs or integrations with in-hours AI systems.
Stroud: This is not a new issue! The semiconductor industry has been wrestling with intellectual property protection and securing the design-to-production ecosystem for years. The challenge is how to build enough flexibility in the ‘fixed’ silicon that, when combined with software (across all layers), is able to guard against future exploits and vulnerabilities. It’s almost impossible to build a modern chip without multiple integrated security capabilities. Also, it’s worth noting that security has to be a multidimensional approach in this age of hyperconnectivity, spanning seamlessly from cloud to edge. This is why we see an ever increasing number of emerging security standards that apply to both implementation and development processes, impacting hardware, software, and system design and deployment.
10: Future Outlook
Q: What do you see as the most important technological and market shifts that will define the semiconductor industry five to ten years from now? How can companies position for sustained leadership?
Bennett: 1) Semiconductor Technology: Chiplets, and the packages that are needed to realize their promise to alleviate the decline of Moore’s Law. 2) Companies: very different answer–the companies that will succeed in the future are those that completely obfuscate the hardware considerations from their customers—it’s all software, don’t worry about the hardware – we have taken care of that.
In summary, in some ways it’s the same old story – recognize and reward the unique engineering talent that helps differentiate your product, understand what the customer wants, and remove the barriers to growth. Sounds simple, right?
Rush: With AI, the amount of data that companies will manage is going to increase tremendously. Trying to manage that traceability is going to be extremely challenging. Jama Connect, with the new scaling improvements and AI vision, is at the forefront of the market and uniquely positioned to help semiconductor companies here.
Gregory: Agreed. AI is already reshaping the demand side of the market equation. The supply-side will evolve to support highly customized semiconductor design, even purpose-built and assembled solutions that are rapidly defined and fabricated. Edge AI and NPUs (neural processing units), along with open architectures such as RISC-V (and the RISC SW Ecosystem), will further broaden the horizons for semiconductor companies. How to be positioned for success? Again, it’s all about response flexibility. Sensing both strong and weak signals in the market and systematically building resilience into the company’s organizational practices will determine which companies emerge stronger from the challenges of the next five to ten years.
Impact Analysis: The Key to Proactive Change Management Success
When consulting with clients, I often convey that there are two types of change management in product development:
Proactive Change Management
Reactive Change Management
Suspect triggers and suspicion are great examples of “reactive change management.” Something changed upstream, and you are notified so you can react.
You may ask, “Mario, wouldn’t it be ideal to react and prepare for change BEFORE it happens?” I would then shake your hand, nod my head in proud agreement, and we would be off to enjoy a festive beverage together.
This describes proactive change management, often referred to in requirements management by its function: “Impact Analysis.” When you take the time to build proper trace links across your requirements, you gain a view of all downstream impacts BEFORE you make a change.
It effectively allows you to notify your teams to prepare for the change and provide details so that when it happens, you can reduce risk and maintain compliance.
In the “old days” of the 1900s, you would handle this by calling all your cross-functional team representatives into a conference room and getting their sign-off. Hopefully, they were paying attention and not on their BlackBerrys or Palm Pilots.
In the modern world, impact analysis is essentially the click of a button, showing you all related downstream items, multiple levels deep—including verification and validation.
Collaborative features such as “discussions,” “subscribing,” and “review and approval” allow for formality in this process, collaboration, and official sign-off (with audit history). For significant changes, this gives teams time to discuss and prepare.
When I work with clients and there are features we are building internally that I know will be useful for them, I often “subscribe” myself to the relevant requirements. This way, if there is an update or status change, I automatically get notified via email.
This keeps me connected to the development process without even having to go into a tool. If I want more information, I simply click on the link and log in.
The Takeaway:
Suspicion catches the change after the fact, forcing teams to react. Impact Analysis allows you and your teams to PLAN for a change BEFORE it happens.
Build strong traceability, accept that change is inevitable, and take a proactive approach to your requirements management change process.
