As we move into 2026, the noise surrounding technology continues to accelerate, with the emergence of new AI models, automation breakthroughs and promises of transformation. But the fundamentals of long-lasting change haven’t shifted. They never do. I’ve compiled what I believe will be the key themes for next year. This is not a view of technology, but a view on what will drive technology adoption.

Transformation is still slow, structural and complex and too often not successful. It demands organizations that can absorb volatility, adapt with purpose, build trust at every layer and execute against the KPIs that matter to leadership, i.e., cost, growth, risk, reliability and time. These aren’t abstract; they are the underlying forces shaping the predictions that follow.

We do need to accept that there is an AI bubble. Model innovation and capital investment have run ahead of the basic conditions that make change stick, such as governance, the ability of people to absorb new ways of working and the organizational muscle to turn technologies into profitable and repeatable business models. Boards are expecting change at a pace that risk, compliance and operating models cannot match. Teams are being asked to adopt copilots and new AI-driven workflows faster than they can be trained, supported or measured. Many AI strategies are, in practice, portfolios of disconnected bets whose unit economics are still unproven.

Real value is being created, but it is unevenly distributed and often obscured by promotional noise. That is what makes this a bubble: not that AI is fake, but that expectations, valuations and narratives have outpaced the practical capacity of most organizations to execute. How and when that gap closes, whether through consolidation, correction or disciplined maturation, is still unclear. Organizations that succeed in 2026 will be the ones that take these fundamentals seriously, not as slogans, but as the quiet architecture behind every technology decision they make. We will succeed if we map that.

1. AI moves from experiments to the enterprise operating system

2026 is the year AI stops being a series of showcases and becomes part of the organizational machinery. The shift is from impressive to integrated. This means:

• AI systems must behave predictably and safely under real conditions and this needs to be demonstrated.
• Integrations must be repeatable and governed, not one-off prototypes that work at 3 AM in an inherently chaotic environment.
• Data quality, lineage and semantics become mandatory, not optional, which expands across the structured and unstructured data to content, emails, documents in SharePoint, etc.
• Models require lifecycle management, not just deployment – operational costs will increase. Transformation will become continuous.
• AI value is measured against operational KPIs, not sentiment, which the CEO is interested in. Productivity is not a measure of outcome; it’s a measure of throughput.

This is the moment where the enterprise AI operating system takes shape, quietly, practically and with discipline. However, for some organizations, the challenge is not scaling their POCs; it’s starting them. Legacy technology, legacy skills and legacy governance are holding many organizations back.

2. Automation strategy becomes layered and context-aware

Automation in 2026 isn’t about replacing processes wholesale. It’s about applying the right tool to the right part of the workflow. Enterprises are learning to separate:

• Deterministic automation for stable processes where reliability is paramount
• AI-enabled adaptive automation for variable processes that require judgment or pattern recognition

The real breakthrough will be recognising that both are critical and the ability to orchestrate both within a single, enterprise operating model will define the winners

This requires:

• Clear risk boundaries
• Strong governance as a built-in control plane – across everything, everywhere, every time
• End-to-end visibility of automated decisions, with humans empowered and skilled to make decisions
• Standardised platforms that reduce fragmentation

Automation becomes a capability that requires continual optimization; it is not a one-time effort.

3. Data architecture becomes the gatekeeper of progress

No matter how strong the ambition surrounding AI and automation, the data estate remains both a constraint and a catalyst. 2026 will make this unavoidable. Organizations that move fastest will have:

• Unified, governed data architectures that treat both structured systems and content, documents, team chats, email, files, knowledge bases and other unstructured signals as part of a single estate
• Shared semantic layers and graphs that eliminate interpretation gaps, clean data is critical and understanding the data is mandatory
• Real-time or event-driven data where operational speed matters, which will drive industry applications
• Full lineage, observability and access clarity, not just internally but up and down the value and supply chain
• Domain, level ownership and accountability are built into the architecture

Data is where resilience and trust manifest technologically: not as theory, but as an engineering reality across both transactional systems and the everyday content people create and use to get work done.

4. Governance turns into the enterprise accelerator

For years, governance was treated as a brake. However, 2026 marks a turning point: governance becomes an integral part of the fabric that enables safe, fast and repeatable deployment. This shift includes:

• Policy-driven decisioning enforced by platforms, not documents. Does every demo we build have this feature built in, or will we add it later (if needed)?
• Continuous monitoring and auditability for AI, data and automation.
• Clear guardrails that give teams confidence to move quickly and skill them to use them.
• Risk, weighted controls that adapt depending on the sensitivity of the workload.
• Integrated security and resilience controls protecting identities, data and workloads by design so that the organization can withstand failure, attack and disruption.

Governance will no longer be seen as a hurdle. It becomes the engine that lets innovation scale without destabilising the organization—security and resilience shift from being specialist concerns to shared properties of how the enterprise operates. Trust becomes the outcome. Acceleration becomes the effect.

5. Workforce and operating models adjust to reality, not predictions

The loudest forecasts predicted rapid workforce displacement due to AI. The real picture in 2026 is more nuanced and far more human. What’s happening instead:

• Organizations that downsized aggressively on the back of AI promises are now quietly rehiring
• AI augments work more often than it replaces it and will begin to activate new business models and workflows
• Copilot experiences and AI assistants are changing how work is done — shifting effort from manual production to review, judgment and orchestration across documents, communications and workflows. They are also changing what works gets done
• New roles emerge around orchestration, oversight and quality assurance. New skills will focus on creativity, critical thinking, imagination and experimental design
• Talent becomes a blend of capability building, reskilling and new operating rhythms
• Skilling and reskilling at scale become the primary levers to empower people to use AI responsibly and effectively, rather than working around it
• The gap between utilising AI and reaping the benefits of AI is primarily organizational, not technological. This is important: breakthrough customers are using the same technology, models, etc. So, what sets them apart?

Execution becomes the shared responsibility of business and technology; tightly integrated and measured on outcomes, not activity. The organizations that thrive will treat workforce transformation and skills as core infrastructure for their AI strategy, not as optional training plans bolted on at the end.

Bringing the predictions together

What defines 2026 isn’t a single breakthrough or disruption. It’s whether organizations have built the underlying capabilities to absorb change, adapt effectively, operate with trust and execute with discipline. AI, automation, data, governance and workforce models only create advantage when they are designed to work together as one operating system, rather than as a set of disconnected projects. The real test is not how advanced any individual pilot appears, but how reliably the entire organization can translate intent into outcomes.

Operational rigour will be fundamental as organizations integrate AI into their business models. This means managing legacy infrastructure, scaling resources intelligently to meet shifting demand, migrating to the cloud, modernising the application stack and controlling costs. Success in 2026 will hinge not only on technological innovation but also on the ability to harmonise modernisation with reliability, ensuring investments in automation, analytics and talent are both strategic and cost-effective.

In practice, 2026 will expose which organizations have quietly invested in this architecture of execution and which have relied on slogans and isolated proofs of concept. The ones that win will be those that treat discipline, safety and readiness as competitive assets, not constraints on innovation.

2026 won’t reward the boldest predictions — it will reward the organizations disciplined enough to execute.