2026 is shaping up to be a crucial year for enterprise AI adoption.
Enthusiasm remains high: According to MIT Technology Review Insights’ recent “Building a High-Performance Data and AI Organization” report, 65% of organizations have already deployed GenAI. Now, organizations are highly focused on harnessing the power of AI to deliver tangible results for their businesses.
When speaking to customers and business leaders across different industries, the priority is to build unified, governed data estates that can power high-quality AI agents and applications. And as companies look to increase the use of these specialized agents and apps that can reason in their unique environments, customized assessments are proving vital.
So what’s next? Here are the trends we predict will shape data and AI efforts in 2026.
Model choice is a non-negotiable
The current battle for supremacy among marginal LLMs has been a boom for enterprises.
AI labs keep pushing each other to make the underlying models more powerful, and organizations don’t want to commit to one provider for fear of missing out on the latest and greatest. Instead, they want the ability to choose LLMs based on their performance and cost for specific tasks.
“When innovation is so fluid, IT flexibility and the ability to switch between underlying models become key competitive advantages. Open technologies provide companies with the control they need to thrive in a new era of sustained AI-driven disruption.” – Dale Williamson, Field CTO
Unified AI governance is critical for enterprise AI agents
Governance is an important layer in agentic AI systems, once only access control is considered.
Governance now extends to AI workloads, dashboards, and more – including semantics and lineage. In short, governance is how organizations control their AI agents. It acts as a contextual layer guiding AI agents to the right data and controlling the system from acting inappropriately.
“Any successful AI strategy needs to answer three questions: Can the business identify the data used? Do they understand which LLMs are being called? And can they explain what happened across the entire agentic AI chain? A strong and integrated governance is key to addressing each of these challenges.” – Robin Sutara, Field CDO
Where does AI development integrate? All data remains
In many organizations, AI development is often divided across potentially dozens of different tools and domains. This impacts overall performance, slows the path to value, and makes it harder for organizations to track and control their AI workloads.
Instead, when companies build AI agents and applications that connect all their data in open and interoperable formats, they eliminate this operational complexity, while also accelerating the pace of AI adoption. Integrated, multi-modal data – spanning structured and unstructured – is the key to success. And with core requirements like unified governance and end-to-end lineage built into the foundation, enterprises can more confidently extend access across their organization.
“The best, most adaptable businesses are using data to navigate a rapidly changing global marketplace. Simplifying AI architectures and building new agents and applications where core, multi-modal business data already exists helps larger numbers of users access this vital, business-critical information faster.” – Dale Williamson
Focused on “boring AI” with human expertise
While some continue their quest for AI superintelligence, enterprises will focus on applying AI to their most repetitive and routine tasks. And they will aim to equip their domain experts with highly specialized AI agents to make the most of their decades of industry experience. Ultimately, the power of AI is about unlocking the potential for innovation for people.
“A people-first approach to AI deployment is critical. Organizations can maximize institutional knowledge by equipping veterans and newcomers alike with specialized tools that keep them focused on high-value tasks.” – Robin Sutara
To learn more about how leaders are accelerating AI initiatives with confidence, read the new MIT Technology Review report: Building a High-Performance Data and AI Organization.
