Futurum Intelligence provides critical insights into digital transformation, focusing on adoption, innovation, and disruption. Backed by a team of industry experts, we deliver research through personalized analyst-client interaction and client portals with visualization dashboards and qualitative and quantitative data reports.
Our research is organized into practice areas aligned with key digital transformation topics, addressing critical business questions. Each area includes analyst coverage and planned deliverables for the year. Through collaboration and strong industry relationships, we identify emerging trends early, helping clients make informed business decisions.
Generative AI represents a platform shift, and when shifts like that happen, they have consequences up and down the technology stack, as well as in areas such as the nature of work, regulations, and, in the case of AI, environmental sustainability. To that end, the Futurum AI practice area covers the core elements of AI while liaising with other Futurum practice areas to reflect that AI spans chips to applications and everything else in between.
The AI area is moving so fast that issues that will be important in 12 months may not even be considered at this point. Nevertheless, organizations need to reflect on what might be coming, and our research aims to support vendors, users, and investors in making the right decisions. As such, these are key coverage areas for the AI practice in the coming 6-12 months.
Agentic AI – the concept of semi-autonomous agents making decisions and taking actions – is still nascent. Still, in 2025, the concept and the agents will start disrupting the traditional application software industry. Vendors that sell large multi-purpose application suites will find that, increasingly, users are not directly interacting with their interfaces; instead, they are using agents to do that on their behalf. This could have all sorts of implications, including, but not limited to, a possible shift from product licenses to agent-usage-based pricing, which might reduce the need for dedicated integration platforms and middleware or result in consolidation of aspects of the software market, such as customer service automation and essential productivity tools. Of course, the application vendors are not sitting idly by and watching this happen; they are developing their agentic platforms. However, the question remains: Do companies want to use a range of different agentic platforms orvhave one sit above all their significant applications, abstracting away the complexity of all the application and data integration points for them?
Large language models are the source of the Gen AI revolution, starting with ChatGPT. Still, while better, more performant models of various sizes will continue to be released regularly, more emphasis than ever will be placed on models of other modalities. These include images, video, audio, and code and modalities prevalent in chemistry and biology, such as chemical structure generation, prediction, protein folding, and molecular design. These might include time series data in financial pattern analysis and process ad workflow optimization. Although large models tend to grab headlines, they are smaller models that enterprise and service providers will increasingly look to as they aim to develop and use models that perform narrower tasks and thus don’t need to be trained on the entirety of the Web. Areas we expect to see traction here include edge computing use cases such as local voice assistants, IoT device control as well as privacy-sensitive domains such as healthcare data processing, latency-critical applications such as gaming, trading systems, and industrial control systems, as well as hybrid architectural issues such as local preprocessing before moving to the cloud and small models for triage and routing.
GenAIOps, the tools and frameworks that address the unique challenges of deploying and managing generative AI systems in production environments, will evolve. These are partly technology changes but also responses to market forces. We suspect that, at some point, solutions will be consolidated into platforms, and the large cloud providers will expand their GenAIOps
offerings. Industry-specific GenAiOps tools could emerge, though it might be too early. There will be more focus on multi-model and multimodal management. On the technical side, there will be more sophisticated prompt management and versioning, automated testing of LLM outputs at scale, and better tools for managing model drift and other aspects in RAG systems. There will be challenges to address, though, including cost management at scale integration with ITOPs, security, and AI governance tools and ensuring consistent performance across deployments.
Companies have been experimenting with generative AI for two years, but such experiments don’t necessarily lead to successful implementations. What is often lacking is a way to ground models in prosperity enterprise data held in siloed, on-premises systems; they will not all simply upload it all to their favorite cloud platform of choice. Therefore, data-related issues will continue to cause challenges and require solutions. Major issues include data quality and governance, including Automated data cleanup and validation, Synthetic data validation and verification, and data lineage for training purposes. Emerging challenges include multi-modal data management, data sovereignty issues, and open-source data quality. At the data architecture layer, issues such as vector database scaling and RAG system optimization will be prominent.
Futurum covers a broad range of AI-enabled consumer and commercial devices. This includes PCs and peripherals, tablets, mobile handsets, XR, hearables and wearables, and IOT, IIOT, and automotive segments. The expansion of AI training and inference from a cloud-centric model to a more hybrid edge-to-cloud model is driving a rapid transformation across the devices segment of the tech stack. With next-gen AI-capable PCs capable of handling 13 billion-parameter models, mobile handsets capable of handling 7 billion parameter models, and wearables pushing against the 1 billion parameter mark, large language models (LLMs) and large mixed models (LMMs) are increasingly moving away from thermally expensive cloud-based silicon to the more thermally efficient silicon powering AI-capable consumer and commercial devices.
The next 6 to 12 months will see an acceleration of AI capabilities in devices. This transition will disrupt core device segments and the entire technology ecosystem around them as cloud service providers (CSPs), independent software vendors (ISVs), and their partners adapt to the emerging hybrid, interwoven AI services model.
Across the technology landscape, vendors are increasingly leaning into their indirect go-to-market strategies and fostering more ecosystem partnerships. There are several drivers for this changing mindset. Economically, vendors are scrutinizing their cost structures more than ever as we have moved away from near-zero interest environments. Companies can no longer justify ‘rampant-hiring’ in their sales & marketing divisions as a GTM tactic. Meanwhile, from a technology standpoint, customer IT environments are getting more complex, residing on multiple clouds and multiple architectures, while the
application landscape is becoming increasingly customized. In short, no technology company can service a customer’s IT needs entirely through its portfolio; partnering is the only way to close the gaps.
Futurum will explore the discipline of partnering in the technology space amidst these trends and future disruptors such as AI. We will explore how GTM is evolving horizontally (e.g., partnering with other technology stacks) and vertically (e.g., embracing an ecosystem of partners engaged in product deployment and service offerings).
Futurum provides hard-hitting coverage for IT leaders that focuses on the evolving role of the chief information officer (CIO) in today’s technology-driven business landscape. As tech becomes more pervasive, CIOs are not just stewards of IT infrastructure but key players in driving strategic innovation, business transformation, and data-driven decision-making. The surge of artificial intelligence (AI), cloud computing, data management challenges, and increased pressure to ensure security and compliance place CIOs at the forefront of the most critical business decisions. Futurm’s coverage delves deep into these trends, providing actionable insights on how CIOs can successfully navigate the complexities of a digital-first world.
A significant focus of our research is the rising demand for CIOs to master the integration of AI and advanced analytics. AI is no longer a distant technology but a driving force behind operational efficiency, customer experiences, and even new business models. As organizations urgently need to leverage AI and cloud infrastructures, CIOs grapple with the dual challenge of scaling these technologies while ensuring ethical, secure, and compliant implementations. Our reports offer deep dives into how CIOs can lead AI strategies, align them with business goals, and avoid the potential pitfalls of poorly managed deployments.
CIOs today must also navigate the complexities of multi-cloud environments and evolving IT architectures, which can either drive agility or result in cost overruns and data silos. Futurum’s CIO Insights helps IT leaders anticipate and respond to these challenges by providing strategic guidance on vendor selection, cloud orchestration, and optimizing hybrid IT ecosystems. Our analysis equips tech companies with a clearer understanding of what CIOs need to succeed. It offers actionable recommendations that help enterprises future-proof their technology stacks, improve resilience, and drive sustained innovation.
Finally, CIO Insights connects with hundreds of leading CIOs around the world and regularly gathers data on key trends, spending patterns, and market shifts. This data is available in our research and in Futurum Intelligence.
Leveraging AI for competitive differentiation and transformation. AI is a key differentiator in 2025, with CIOs under increasing pressure to use it not only for operational improvements but also to create new business models and revenue streams. Success in AI deployment will hinge on how effectively CIOs align AI with broader digital transformation strategies, ensuring it drives true competitive advantage.
Ensuring cybersecurity resilience in an evolving threat landscape. As cyber threats become more sophisticated, CIOs in 2025 must prioritize building cybersecurity frameworks that are both resilient and adaptable. With the rise of distributed workforces and cloud-based infrastructures, securing data and maintaining compliance while fostering innovation will remain a top challenge.
Futurum covers a broad spectrum of cybersecurity-related technologies, including application, cloud, data, endpoint, network security, identity and access management (IAM), and integrated risk management and Security Operations Center (SOC) markets. With this in mind, our vantage point spans key use cases, including threat hunting and intelligence, incident response, attack detection, and cyber-recovery. Key themes of our coverage include the modernization of the SOC and the adoption of technologies that intrinsically enhance cyber-resiliency for data, applications, and infrastructure.
Malicious actors are using AI to develop more evasive and sophisticated threats with greater speed and efficiency. In response, defenders must also adopt AI-powered tools to reduce the likelihood of a successful breach and mitigate the resulting damage in the event of a successful breach. Vendors are responding in kind by baking AI-based capabilities into their solutions to support security and infrastructure teams alike from the standpoint of their cyber-resiliency.
Attack surfaces are sprawling and becoming more splintered than ever, thanks to the rise of remote work, the Internet of Things (IoT), and multi-hybrid cloud IT and application environments. This is compounding the fact that threat vectors are changing at an unprecedented pace and becoming more sophisticated than ever, as discussed in the prior bullet. More than half of organizations plan to add a new cybersecurity vendor to address these changing requirements, and 45% plan to add a new cybersecurity product category in 2024, according to The Futurum Group’s Cybersecurity Decision Maker IQ data.
The network of cybersecurity tools typical enterprises use is overwhelmingly large – with organizations typically counting their tools in the dozens. The need for faster reactiveness to next-generation vulnerabilities and attacks, coupled with the need for security and IT operations teams to scale across larger and more complex application and IT environments, is driving the need for a more consolidated approach to streamline operations, reduce protection and detection gaps, and to accelerate threat response. That being said, decision-makers will be mindful of relying on fewer vendors – the risk of which was evidenced by the July 2024 CrowdStrike outage.
With data constantly under threat, organizations require far more than traditional snapshot, replication, backup, and operational recovery capabilities. Vendors are responding to facilitate a more preventative stance against attackers and minimize data loss and downtime of critical business services following an attack. For example, they are baking in and acquiring capabilities for detecting potentially malicious activity, incident response, and data security posture management (DSPM).
We are experiencing one of the greatest periods of innovation in creating and utilizing software. Once considered a back office expense, business strategies rely heavily upon their technology organizations and partners to rapidly deliver innovations, competitive capabilities, and compelling digital customer experiences through software. DevOps and Application Development are crucial practices encompassing end-to-end aspects of creating and operationalizing software across the Software Development Lifecycle (SDLC).
DevOps is not static; it continues to evolve and adapt to the changes in how we create software and the technologies used in contemporary software and infrastructure stacks. This evolution is evident in the development pipelines that span containers, container orchestration, microservices, distributed architectures, cloud, multi-cloud, on-prem, and legacy applications. The principles of DevOps and Agile have not only adapted but also led to the creation of new disciplines, including system reliability engineering (SRE), platform engineering, DataOps, AIOps, SecOps, GitOps, ITOps, FinOps, and more.
Security and AI/ML/GenAI are as much a part of software creation as writing, building, and testing software. DevSecOps continues to mature and expand its scope across the SDLC. AI-based copilots, code generation, QA planning and testing, and agentic and guided AI agents are reshaping the technologies and workflows we use to create, deliver, secure, and operate software.
Technology and business leaders must track advances that can aid them in making sound decisions that fulfill today’s needs and deliver on their organization’s strategic business objectives, which rely on software.
Enterprise Applications are the lifeblood and framework for accomplishing work in the modern organization. We examine 12 categories of applications used in the enterprise, including CRM, ERP, Project Management, Productivity, Collaboration, CCaaS/UCaaS, Digital Transformation/Data and Process Management, Business Intelligence/ Analytics & Visualization, Data Management/Platform, HR/Employee Experience/Rewards. Content Management, Line of Business, Marketing Automation, and SCM, and delve into how they shape the broader enterprise information architecture. We also focus on the underlying technologies and systems that power these applications, including artificial intelligence and automation, and assess how employee engagement and experience trends impact the market.
In under two years, generative AI has progressed from a novel new technology with few defined use cases to an integral part of many enterprise applications and platforms. Including the tech has not led to significant productivity or efficiency gains yet. Still, the arrival of AI-powered agents will likely be the catalyst for driving substantial ROI. The challenge will largely be around prioritizing the accuracy and efficiency of these agents and generating trust among those responsible for business outcomes and everyday workers being asked to interact with these tools.
The rise of generative AI also impacts the expected pricing models and approaches of application vendors. Greater efficiency
and the lower cost of using automation is driving a reassessment of traditional seat pricing models. New approaches, including
consumption-based and outcome-based pricing, are being floated and tested by the market. It is still early for new models, but the market will likely rely on something other than expanding seat licenses to drive revenue growth.
The proliferation of applications purchased and used by enterprises ballooned over the past several years, in part due to the
need to quickly stand up new features during the COVID-19 pandemic but also in response to the desire to digitally transform
all facets of a company’s operations. The result is often a mishmash of applications, often with several applications performing
similar functions and some simply dormant applications, creating a combination of security issues, economic waste, and
technical debt that organizations must address. As pricing models shift, we can expect a further inspection of the enterprise
tech stack, which may lead to market share shifts.
Organizations operating within specific industry segments have traditionally selected point solutions from vendors that were entrenched in their sector and, as a result, built applications that took into account the specific and nuanced features, workflows, and regulatory requirements that accompany each distinct industry. But as larger enterprise platforms seek to gain market share, they’re increasingly using their might to release industry-specific versions of their platforms and applications, often loaded with the latest features and functions, promises of frequent updates, and greater flexibility. Vendors of all stripes have realized the opportunities created by scaling and segmenting their product offerings to meet the distinct needs of these industry-specific customers.
Ultimately, enterprise applications and how they are deployed, consumed, and priced are at a crossroads, thanks to the rapid
ascension of generative AI and the changing nature of work. Vendors will be focused heavily on leveraging generative AI and
integrations to help position their respective platforms as the center or hub where enterprise work is conducted.
For more than 50 years semiconductors have been the enabling technology that have transformed the global economy and society by delivering innovations in Computing, Telecommunications, Consumer Electronics, Automotive, Industrial Electronics, and Military and Civil Aerospace. The semiconductor market has grown to over $600 billion in annual sales (in 2024) and, as the era of Artificial Intelligence (AI) begins, annual sales exceeding $1 trillion are expected within a decade as the semiconductor industry once again plays a critical role in delivering technological advances.
Semiconductor Industry Cycle:
The semiconductor industry is highly cyclical with revenue growth heavily influenced by the timing of step function increases in manufacturing capacity set against ongoing, incremental increases in demand. Demonstrating its volatility, semiconductor chip market growth of about 20% in 2024 will give way to modest growth of less than 10% in 2025, as supply and demand comes more into balance. Spending on semiconductor manufacturing equipment is estimated to have grown by about 7% in 2024 and, as the industry attempts to pace investment in manufacturing capacity with market demand, another year of modest, single-digit growth is expected in 2025.
End Market Drivers:
The drivers of semiconductor market growth are shifting away from the traditional high-volume but lower growth mainstream markets, such as Cell Phones and PCs, towards more premium, higher growth markets defined by the high-performance technologies, such as AI Processors and High Bandwidth Memory (HBM), required by Data Centers and Autonomous Vehicles, and the specialist technologies, such as compound semiconductors, used in Electric Vehicles.
Semiconductor Manufacturing:
Demand for advanced node foundry capacity is expected to continue apace, driven by the needs of high-performance computing (HPC) applications. Semiconductor manufacturers will accelerate the move to sub 20nm node production for leading edge chips and will begin readying 2nm process technology for mass production in the most advanced fabs. Mainstream and mature node manufacturing will continue to recover and soak up excess fab capacity for greater than 20nm node production, driven by China’s self-sufficiency strategy and ongoing demand from a wide range of end markets, including Automotive and IoT.
Technology Roadmap:
For decades, Moore’s Law delivered performance and cost improvements through shrinking transistor geometries, increased die sizes and larger diameter wafers.
However, as Moore’s Law slows down, the industry is expanding investments in “More Than Moore” innovations such as:
And in Emerging Technologies, such as:
Geopolitics: Global political tensions and disruptions are influencing the semiconductor industry structure and dynamics more than ever as governments around the world increasingly recognise that the industry is a vital national security asset. Policies designed to mitigate the risks associated with globalization (such as supply chain disruption) are being rolled out, for example:
Regulatory Environment: In recent years, the global trend in developed economies has been towards “big government” and a tighter regulatory environment. Initiatives such as “DEI” (Diversity, Equity and Inclusion and “ESG” (Environmental, Social and Governance) and “Net Zero”, load up the industry with increased compliance costs and divert funds away from investment in productivity-enhancing innovations. In the unfolding era of AI, as semiconductors are becoming an even more important enabling technology, helping to deliver solutions to problems across society, environmental regulations that are too onerous and lack of access to cheap and reliable sources of energy are of particular concern to semiconductor industry leaders.