Industry executives and experts share their predictions for 2026. Read them in this 18th annual VMblog.com series exclusive.
By Claus Jepsen, Chief Technology Officer, Unit4
As we race to AGI, businesses will derive more value from expanded AI capabilities in 2026, and IT leaders will prepare for market and regulatory volatility.
As 2025 draws to a close, the possibility that the AI bubble will burst soon is on everyone’s mind. Google CEO Sundar Pichai admitted as much in a recent BBC interview, saying that there are “elements of irrationality” in the massive investment in AI infrastructure.
Pichai also said “no company would be immune” if the bubble bursts, including Alphabet. It’s fair to say many investors and business leaders are anxious about what 2026 will bring, but there’s also reason for optimism about how the new year will play out.
In 2026, we’ll see value created as a byproduct of the drive toward AGI, such as AI-driven abstraction that advances composability in enterprise software like ERP. We’ll also derive value in the meaning provided by metadata and context-powering AI agents.
Here are three predictions for 2026 that touch on how AI-infused business software will evolve to deliver new value and how IT leaders will prepare for a volatile market and more complex regulatory environment in the months to come.
1. AI-driven abstraction will transform enterprise systems, advancing composability.
The products the leading AI companies have delivered so far are overhyped, but the foundational models contain aspects that are transforming enterprise systems by making it simpler to access information without creating APIs and user interface tools to extract or push data. This advances the goal of composability, which in practical business terms, means the ability to simplify enterprise software interactions across modules.
The goal is a conversational experience for users, and AI-driven abstraction functions as a glue that binds disparate systems together so that users can extract or push data without the technical challenges of building and maintaining the “wiring” that enables the flow of data across multiple systems with different protocols.
The ability of LLMs and AI agents to simplify interactions will be transformative, not just for ERP, but for all enterprise software. For users, composability will drive greater efficiency and enable more flexibility across product sets. But it’s a challenge for software developers who will have to demonstrate value when their core product is largely invisible to users.
2. Metadata and context will create a new source of value.
AI agents require rich ontologies and contextual information in their subject areas to function effectively. Concepts, properties, rules and relationships within the domain must be clearly defined so the agent can respond to questions, in the same way that a human needs subject matter expertise and data to answer complex questions.
Think of what happens behind the scenes when a business user asks an AI agent a question like, “Are our suppliers at risk?” The agent must first understand what a supplier is and who the company’s suppliers are. It needs access to relevant specific data like invoices and accounts receivable, as well as general information on the market.
Richer metadata and context make the AI agent more capable and enable it to add value that transcends the system or record from which it draws knowledge. It’s the meaning derived from defining those ontologies and providing context that becomes a source of value in itself. This will be critical for smart, context-aware enterprise solutions.
3. IT Leaders will focus on cloud sovereignty, strategic risk management and automation in 2026.
In the coming year, geopolitical dynamics and tension between laws in the U.S. and privacy protection measures in European countries will result in a greater focus on cloud sovereignty and strategic risk management. There is also a rising appetite for automation as businesses look for ways to increase operational speed.
IT leaders have always been concerned about how and where data is stored, but passage of the US CLOUD Act gives the U.S. government the authority to demand customer data from U.S.-based tech companies, regardless of where the data is stored. This creates conflicts between U.S. hyperscalers used by companies that also must comply with EU privacy laws. The tension is leading companies to build alternatives in Europe and consider cloud mobility as a risk management strategy as they navigate choppy regulatory waters.
Also look for a more intense focus on automation in 2026 as businesses look to increase the speed of traditional workflows. The focus on automation isn’t necessarily related to AI. Demand for automation in general is rising as businesses seek efficiency and look for an advantage in the coming year.
##
ABOUT THE AUTHOR
Claus Jepsen is the Chief Technology Officer at Unit4, where he is building cloud-based, super-scalable solutions and bringing innovative technologies such as AI, chatbots and predictive analysis to ERP.
