16 April 2025
AI as a Foundation: Deloitte’s 2025 Tech Trends highlight AI as a foundational element across enterprise technology, likened to electricity in its ubiquity.
Spatial Computing Integration: AI enhances spatial computing (AR/VR), enabling immersive, adaptive environments that respond to user behavior in real-time.
Shift to Agentic AI: The focus is moving from large language models to autonomous AI agents capable of executing tasks and workflows independently.
Hardware Evolution: The demand for AI is driving a hardware boom, with AI-specific chips and AI-enabled devices becoming essential for performance and privacy.
AI-Augmented IT: AI is transforming IT operations, from code generation to predictive maintenance, amplifying efficiency and reducing manual workloads.
Quantum-Resistant Security: With quantum computing on the horizon, there’s an urgent push towards post-quantum cryptography to safeguard data.
Intelligent Core Systems: AI is being embedded into core enterprise systems (like ERP and CRM), enabling real-time, autonomous decision-making and operations.
Deloitte’s Tech Trends 2025 report spotlights how artificial intelligence (AI) has become foundational to the latest 2025 technology trends in business and IT. In fact, the 16th annual report emphasizes that AI is now a “common thread” running through nearly every major trend and is poised to be part of “the unseen substructure of everything we do”.
In other words, AI is expected to be as ubiquitous and essential as electricity, quietly powering experiences in the background. From optimizing city traffic and personalizing health care to enabling adaptive education, AI will increasingly work behind the scenes to make systems smarter and more intuitive“like magic, but grounded in algorithms”.
While generative AI dominated headlines as the buzzword of 2023, Deloitte’s analysis makes clear that the technology trends shaping 2025 extend beyond any single AI capability. Instead, myriad forms of AI are being woven into the fabric of how we interact with computers, how information is processed, and how infrastructure is built.
For enterprises, this means AI is no longer a standalone project or novelty. It has become integral to nearly every emerging tech strategy, product, and process.
Below we explore how AI underpins each of Deloitte’s six key Tech Trends 2025 categories: Interaction, Information, Computation, Business of Technology, Cyber & Trust, and Core Modernization.
In each area, from spatial computing and next-generation AI models to new hardware, IT operating models, cybersecurity, and core systems AI’s role is foundational. We supplement Deloitte’s insights to validate and enrich the picture of these trends in technology.
Source – Deloitte
Deloitte’s 2025 Tech Trends reveal AI as the unifying force reshaping enterprise technology. Spatial computing is merging with AI to create immersive, intuitive environments. Organizations are moving beyond large language models to task-specific, autonomous “agentic AI.” AI’s demand for computing power is fueling a hardware resurgence, with AI chips projected to exceed $50B in 2024.
Generative AI is revolutionizing IT operations, coding, and delivery models. Meanwhile, quantum computing threats are driving adoption of post-quantum cryptography. Core systems like ERP are being rebuilt with AI-first architectures, evolving into intelligent, self-optimizing agents that automate decision-making and enable real-time enterprise adaptability.
Spatial computing, encompassing augmented reality (AR), virtual reality (VR), mixed reality and digital twin technologies, is taking center stage as a top enterprise trend. Its appeal lies in the ability to merge digital information with the physical world in intuitive ways.
According to Deloitte, spatial solutions can break down data silos and create more natural interactions for both workers and customers. Companies are already finding success using AR/VR for advanced simulations that let them test scenarios and visualize outcomes before making real-world changes.
For example, a manufacturer might simulate a factory floor reconfiguration in VR to see its impact on operations, or a city could model traffic flow changes in a digital twin to optimize congestion management.
Looking ahead, AI is set to play a key role in unlocking the full potential of spatial computing. By layering machine intelligence into these immersive environments, organizations can enable seamless experiences that adapt to user behavior in real time.
Deloitte predicts that ongoing AI advancements will lead to spatial computing platforms with improved interoperability and built-in intelligence, ultimately enabling AI agents to anticipate and proactively meet users’ needs in these environments. Imagine virtual collaboration spaces that automatically bring up relevant data based on conversation context, or AR training goggles that observe a worker’s actions and offer just-in-time guidance. AI can make interactions feel “magical” by handling background tasks and context so the experience is fluid for the user.
This convergence of spatial tech and AI is not just theoretical, it aligns with broader market expectations. Gartner, for instance, projects that the spatial computing/extended reality market will explode from about $110 billion today to over $1.7 trillion by 2033. Much of that growth will come from enterprise and industrial adoption, where AR/VR headsets and digital twin platforms solve specific business problems in design, training, field service, and more.
To enterprise tech leaders, this trend highlights an emerging class of trending technologies that deliver immersive experiences powered by AI insights. Those investing early in spatial computing, and the AI to make it intelligent, could gain advantages in training effectiveness, operational efficiency, and customer engagement.
Generative AI adoption has skyrocketed, a Gartner poll found 70% of organizations in “exploration” mode and 19% already in pilot or production with generative AI initiatives.
The rapid rise of tools like ChatGPT in late 2022 spurred businesses across industries to experiment with large language models (LLMs) and other generative AI applications. Early adopters have seen promising results in areas like content creation, customer service chatbots, and code generation. In fact, Gartner’s survey indicates the vast majority of enterprises are now somewhere on the journey of exploring or implementing generative AI, a testament to how quickly this trending technology has captured executive attention.
Yet Deloitte’s Tech Trends 2025 report reminds us that the future of AI is about much more than just today’s big-model, generative AI hype. Leading organizations are already looking beyond LLMs and rethinking their AI strategies for efficiency and specificity. “Despite their general applicability, LLMs may not be the most efficient choice for all needs,” the report notes, and many enterprises are now considering multiple smaller models tailored to particular tasks.
By training domain-specific AI on smaller, more accurate datasets, and leveraging open-source models and multimodal AI that can handle images, speech, and more, companies can deploy the right AI tool for each job rather than one monolithic model for everything. This shift promises to make enterprise AI development more cost-effective and business-focused.
Crucially, the next chapter of enterprise AI will focus on execution, not just information. Instead of AI that only generates content or answers queries, the emerging vision is AI that can take actions on behalf of users.
Deloitte foresees an “era of agentic AI” in which businesses and consumers are armed with AI co-pilots – autonomous agents capable of carrying out discrete tasks and workflows. As the report quips, the popular saying “There’s an app for that” could soon become “There’s an agent for that”.
Imagine an AI agent that can scan your financial data, identify missed revenue opportunities, and automatically execute certain adjustments, or a marketing AI that not only drafts social media posts but also autonomously schedules them and adjusts strategy based on performance. Such agents would move AI from the back-office to center stage in day-to-day operations.
This trend toward autonomous agents is echoed by other experts. Forrester, for example, has named “AI agents” as one of the top emerging technologies, noting that autonomous workplace assistants are growing increasingly sophisticated and expanding from back-office use to customer-facing roles.
These AI agents are learning to better understand nuance and context, making them more adept at complex interactions. The business implications are significant: companies might deploy fleets of specialized AI agents (for finance, HR, IT support, customer service, etc.) to continuously handle tasks and optimize processes.
Of course, realizing this vision requires overcoming current challenges. Many enterprises are still grappling with how to scale even basic generative AI models securely and effectively. Surveys show a majority of pilots haven’t yet reached production due to issues like data quality, privacy, and model accuracy.
It’s also clear that human oversight, ethical guidelines, and strong AI governance will be needed to guide these powerful tools. Nonetheless, investment is pouring in. Gartner forecasts worldwide spending on generative AI tech to reach $644 billion in 2025, reflecting the race to capitalize on AI’s potential.
The message for enterprise leaders is that AI innovation isn’t slowing down after the initial hype, it’s evolving into more practical, targeted solutions that will deeply impact products, services, and operations. Preparing for a future of many AIs (and AI agents) working in concert should be a priority for any organization’s strategy.
As artificial intelligence becomes pervasive, it is reshaping the underlying infrastructure and hardware that enterprises rely on. Deloitte’s “Hardware is eating the world” trend declares that the AI revolution “depends on access to the appropriate hardware”.
In recent years, we’ve seen a dramatic reversal of the old paradigm that software alone drives innovation. Today, cutting-edge chips – GPUs, AI accelerators, specialized silicon – are strategic assets in the race for AI capability.
A case in point is NVIDIA, once a niche graphics chip maker, it is now among the world’s most valuable tech companies, as its AI-optimized GPUs became the picks and shovels of the generative AI gold rush. In early 2024, NVIDIA even briefly hit a $2 trillion market valuation, driven by insatiable demand for its AI chips across cloud providers and enterprises.
Deloitte’s research quantifies this hardware boom. By the firm’s estimates (based on World Semiconductor Trade Statistics data), the market for chips dedicated solely to generative AI workloads is projected to exceed $50 billion in 2024.
To put this in perspective, that is a massive new segment of the semiconductor industry arising almost overnight, fueled by AI models’ need for parallel processing and high-speed computing. And it’s not just data-center chips, the report points to a coming wave of AI-embedded end-user devices.
For example, PC manufacturers like AMD, Dell, and HP have started touting “AI PCs” with built-in AI co-processors to future-proof enterprise laptops. These AI-enabled PCs can run machine learning tasks locally (for image generation, data analysis, etc.) without relying on the cloud, offering faster responses and enhanced privacy. Early use cases suggest knowledge workers armed with such AI PCs could be “supercharged” with instant insights and on-device automation.
For CIOs and CTOs, the implications are twofold. First, infrastructure is once again a strategic differentiator. Gaining competitive advantage may require investing in specialized hardware or cloud capacity for AI, and optimizing everything from networks to cooling systems to handle AI’s heavy workloads.
Gartner notes that major cloud providers are already pouring money into this. In 2024, nearly 60% of hyperscalers’ server spending is expected to go toward AI-enabled servers. Enterprises should assess how to best obtain the AI computing power they need, whether through cloud partnerships, on-premises AI appliances, or edge devices.
Second, organizations must plan for a more distributed IT landscape. Some AI tasks will run in big data centers, while others might run at the edge or on user devices (for latency, cost, or privacy reasons). Managing this efficiently will be key. There’s also an energy component: training and running AI models can be power-intensive, so sustainability will factor into hardware decisions (for instance, using more efficient chips or cooling methods).
As Deloitte puts it, after years of pushing IT toward virtualization and cloud centralization, AI is sparking a shift back toward investing in hardware and purpose-built infrastructure. In short, enterprises that want to leverage AI at scale must pay close attention to the “nuts and bolts” – chips, servers, devices, and the architectures connecting them. The latest tech trends in AI can’t be harnessed fully with yesterday’s hardware.
One of the most profound impacts of AI’s rise is on the IT department itself, how software is built, and how technology teams operate. Deloitte calls this trend “IT, amplified,” noting that as the tech function shifts from leading digital transformation to leading AI transformation, it’s an opportunity to redefine the future of IT’s role. In practical terms, AI is turbocharging many aspects of IT work:
According to Deloitte, companies that aggressively embrace these AI-enabled ways of working in IT are seeing results. In a recent Deloitte survey, tech companies at the forefront were 2× more likely than others to report that generative AI is already transforming their organization or will do so within the next year.
Forward-thinking CIOs and IT leaders are treating the current AI wave as a “once-in-a-generation opportunity” to reimagine IT’s structure and value. This can mean redefining roles (e.g. introducing AI prompt engineers or data curators), retraining staff, and revising vendor strategies to incorporate AI capabilities. It also means reshaping processes, for instance, moving from traditional software development to an AI-assisted DevOps pipeline, or from reactive IT support to proactive, AI-driven monitoring.
Notably, AI is prompting a shift away from the recent era of ultra-lean IT budgets. After years of cost-cutting and cloud off-loading, organizations are realizing they must invest in talent and tooling to harness AI. Gartner predicts global IT spending will rise to $5.26 trillion in 2024 (up 7.5% from 2023), with much of that growth directed toward new technology initiatives like AI.
In PwC’s June 2024 Pulse Survey of tech execs, 85% said they now have the capability to scale business models using AI, IoT and advanced semiconductors, and 76% plan to boost use of generative AI to drive that scaling. Furthermore, 61% of tech firms have already adopted generative AI in at least a few areas of their business, and 84% of executives are increasing their cloud budgets in the next cycle with generative AI as a leading driver. These figures underscore that IT leaders are investing real resources to embed AI across operations.
With these investments, however, come new challenges. Talent is a big one, AI expertise is in high demand and short supply. CIO magazine’s State of the CIO 2025 survey found 54% of CIOs already feel that staff and skill shortages are hindering their strategic goals, and the most acute gaps are in AI/ML (38% of CIOs), cybersecurity (33%), and data analytics (21%).
As one tech executive noted, “If 40 to 60% of companies embrace these trends, chances are we won’t have enough people to do it”. This means IT leaders must upskill existing teams, partner with external providers, and find creative ways to attract and retain AI talent.
Additionally, CIOs need to guide their organizations through AI governance, setting policies on AI ethics, bias, and risk (an emerging priority identified by Gartner as well. The bottom line is that as AI becomes integral to enterprise strategy, the IT function’s remit is expanding. Tech leaders have a mandate to ensure AI delivers business value responsibly, which requires new thinking in everything from architecture to ethics. Those who navigate this well will elevate IT from a support role to a central driver of innovation and growth.
Amid all the excitement about AI, Deloitte’s Tech Trends 2025 report reminds enterprises not to lose sight of another transformative technology on the horizon: quantum computing. Specifically, the trend “The new math: Solving cryptography in an age of quantum” focuses on the urgent need for organizations to future-proof their cybersecurity in preparation for quantum advances.
The concern is straightforward: today’s encryption protocols, which secure everything from online banking to confidential business data, rely on mathematical problems (like factoring large prime numbers) that would take classical computers eons to solve.
However, quantum computers operate on completely different principles and could theoretically crack these problems exponentially faster. As one security expert put it, a problem that takes thousands of years to break today might be solved in hours on a sufficiently powerful quantum machine.
While viable, large-scale quantum computers are still in development, the threat is not seen as far-fetched. Nation-states and tech giants are investing heavily in quantum R&D, and Gartner estimates that by 2029 most current forms of cryptography will be unsafe to use due to quantum breakthroughs.
Enterprises cannot afford to wait until that moment arrives; data intercepted today (even if encrypted) could be stored and decrypted later when quantum capabilities mature, a risk known as “harvest now, decrypt later.” Therefore, preparation must begin now, which is why this has emerged as a top technological trend in the security realm.
The solution path is post-quantum cryptography (PQC), new encryption algorithms and protocols designed to resist attacks from quantum computers. In 2024, the U.S. National Institute of Standards and Technology (NIST) released a set of vetted quantum-resistant algorithms as new standards.
These methods rely on mathematical problems (lattice-based, hash-based, etc.) that even quantum computers should struggle with. Deloitte reports that organizations have already started the transition: soon after NIST’s announcement, Apple updated iMessage to use a quantum-secure encryption mode, Google implemented PQC in its Chrome browser and crypto libraries, IBM integrated PQC into its platforms, and Microsoft announced similar plans for its products. In addition, a consortium of companies is working with NIST’s National Cybersecurity Center of Excellence on a project to inventory and test PQC solutions.
For enterprise leaders, the mandate is clear. Much like the scramble to patch systems for Y2K (if not bigger), there’s a multi-year roadmap needed to migrate to quantum-safe cryptography. Deloitte suggests first gaining an understanding of where and how your organization uses cryptography today, an inventory of all applications, devices, and services that rely on encryption.
With that in hand, security teams can prioritize which systems to update first and develop a transition plan to new algorithms. This “crypto agility” will be critical; it may involve updating protocols, replacing certificates, and ensuring compatibility across a wide range of infrastructure. Gartner strongly recommends deeper research and proactive planning in this area now, predicting that organizations who delay will face significant security and compliance risks later.
It’s worth noting that this quantum encryption upgrade can yield side benefits too. Deloitte likens it to “cleaning out the basement”, as companies dig into their legacy systems and forgotten corners of IT to swap out crypto, they often discover other issues to fix or opportunities to modernize (such as improving key management, adopting zero trust architectures, or retiring obsolete systems).
In essence, preparing for quantum can become a catalyst to strengthen cybersecurity and trust practices overall. And given that AI and automation are increasingly woven through business processes, ensuring those AI systems operate in a secure, trusted environment is paramount. AI adoption will falter if customers and stakeholders don’t trust the underpinning systems to be safe. Thus, quantum-proofing security is an essential parallel effort alongside AI innovation – a message enterprise tech strategists are heeding now.
The final trend in Deloitte’s 2025 outlook, “The intelligent core,” examines how AI is transforming the central systems that run businesses. The big ERP, CRM, and database platforms are often referred to as systems of record or the “core.”
Traditionally, these core applications have been the authoritative source of truth for all critical business data. If you had a question about inventory levels, customer orders, or financials, you went to the ERP system for the answer. That model is now being fundamentally challenged by AI (DI_Tech-trends-2025.pdf).
Enterprise software vendors are going “AI-first” in redesigning their offerings. This means embedding machine learning and automation capabilities throughout their modules, from procurement and supply chain to HR and finance.
AI algorithms can sit on top of core data and learn an organization’s patterns, workflows, and rules. As a result, users might not need to navigate complex ERP menus or run static reports to get information. Instead, an AI-powered layer can draw from the core in real time and present insights or even take actions via more user-friendly interfaces (like natural language queries or chatbots).
For example, rather than a manager running a report in a CRM system to identify at-risk customers, an AI assistant could proactively alert, “Customer X is likely to churn next month and here are recommended interventions,” pulling that insight from core CRM data plus external signals.
Deloitte says it’s hard to overstate AI’s transformative impact here. This goes beyond automating tasks to “fundamentally rethinking and redesigning processes to be more intelligent, efficient, and predictive”.
Consider planning and forecasting, an AI that continuously learns from historical data and current trends might soon outperform traditional budgeting processes, and do so on the fly. Or in supply chain management, an AI could dynamically reroute shipments and adjust inventory in response to real-time events (weather, strikes, demand surges) without waiting for human decisions. Essentially, the core system itself becomes dynamic and autonomous, rather than a static ledger of transactions.
We are already seeing steps in this direction. Many ERP vendors now offer AI-driven predictive analytics, anomaly detection, and automated decision workflows as add-ons to their platforms. Over time, these may become default features.
Deloitte envisions that eventually AI will evolve from merely informing core decisions to taking direct action – a set of trusted AI agents handling routine decisions within core business processes.
For instance, an AI agent in an “intelligent core” could automatically approve low-risk expense reports, reorder stock when inventory runs low, or reassign workloads among project teams for optimal productivity, all according to parameters learned from the organization’s data and goals. The ultimate endpoint is autonomous core operations, where many day-to-day decisions are executed by AI at digital speed, and humans intervene by exception or to provide strategic guidance.
To reach this point, companies will need to overcome notable challenges. Integrating AI with complex legacy systems is non-trivial – it requires robust APIs, data integration, and often cleaning up the data in those core systems. Data governance becomes even more critical as AI pulls from sensitive enterprise data, firms must ensure accuracy, privacy, and security.
Deloitte warns that as more AI is integrated, architectures become more complicated, and ensuring the AI’s actions are correct and trustworthy (especially in financial or safety-critical processes) is a key hurdle. Change management is also significant; employees who have long used certain systems will need to adjust to AI-driven processes and trust the recommendations or actions of these new tools. There may be initial resistance or a learning curve in shifting roles, for example, an accountant moving from manual reconciliations to supervising AI outputs.
However, the potential benefits are major. Organizations that crack the code of an intelligent core stand to gain agility and speed that competitors will struggle to match. They can operate in something closer to real time.
Deloitte notes that few companies today fully reap the benefits of their large ERP investments, in fact, by 2027 over 70% of new ERP deployments may fail to meet their business objectives. AI could be the key to finally unlocking more value, by making these systems adaptable and smart rather than rigid.
We may also see new business models emerge, as AI-enabled core systems allow for hyper-customization of products, predictive service offerings, and other innovations that were not feasible with static software.
For tech and business leaders, the takeaway is to begin infusing AI into core system upgrade plans now. Many companies are in the midst of ERP modernizations or migrating core functions to the cloud, it’s an ideal time to evaluate AI capabilities in those new systems and pilot intelligent process automation in high-impact areas.
Those who wait might implement a “plain” core system and then find themselves overhauling it again to add AI. Instead, building an intelligent core by design – one that evolves with AI advancements – will position enterprises to be far more competitive and resilient in the coming years.
Deloitte’s Tech Trends 2025 paints a compelling picture: AI is no longer a standalone trend, it’s the foundational fabric enabling and accelerating all other technology trends. From immersive interfaces and smarter analytics to autonomous infrastructure, security, and business processes, AI’s role is both at the forefront and in the background of enterprise innovation.
The enterprises that navigate this shift successfully will be those that embrace intentional intersections, marrying AI with other emerging tech trends and with human creativity to reimagine what’s possible.
As Deloitte aptly asks, what convergence will your organization discover next? In the journey through 2025’s tech trends and beyond, those who weave AI strategically into their core strategies, while keeping an eye on complementary innovations and guarding trust, will lead the way in the next era of digital renaissance.
