Daniel Fallmann is founder and CEO of Mindbreeze, a leader in enterprise search, applied artificial intelligence and knowledge management.
getty
Enterprise AI is entering a more consequential phase. The market is no longer satisfied with generic assistants that can summarize documents, draft emails or answer broad questions with equal confidence for everyone in the company. That model delivered early productivity gains, but it did not fundamentally change how work gets done.
The next phase is defined by systems that understand who the user is, what they are responsible for, what they are allowed to access and what decisions sit inside their role. Enterprise AI is becoming role-aware, context-specific and permission-sensitive, and that shift is already measurable.
According to Pew Research Center, as of October 2025, 21% of U.S. workers report using AI in their jobs, up from 16% the year prior, signaling rapid integration into daily workflows.
This evolution is architectural. Systems that simply respond to prompts cannot support execution at scale without understanding context. MIT Sloan explains that agentic AI systems are increasingly capable of integrating with enterprise software and completing tasks with minimal human intervention, which requires deep awareness of workflows, permissions and user roles. Once AI begins to act within systems rather than simply respond, generic access becomes a liability. Context is no longer optional, but becomes foundational.
In practice, personalization in the enterprise is often misunderstood. It is not about tone, interface preferences or remembering prior interactions. It is about delivering relevant, constrained and actionable information based on role. A compliance officer, a CFO and a frontline employee may ask similar questions, but the correct response depends on their authority, data access and operational responsibilities.
Reuters reported in March 2026 that connecting AI tools to enterprise-wide data sources without strict access controls can expose sensitive internal information, underscoring the importance of permission-aware systems.
This is where governance and personalization intersect. The National Institute of Standards and Technology emphasizes that AI systems must account for evolving identities and permissions over time, including both human users and non-human systems acting on their behalf. That requirement introduces a level of complexity that most early AI deployments were not designed to handle.
Role-based AI cannot rely on static rules. It must continuously interpret context, authority and risk in real time.
The concept of AI as a “personal knowledge hub” is often framed too narrowly. In consumer terms, it suggests convenience. In the enterprise, it must mean something more precise.
A true knowledge hub does not simply retrieve documents. It assembles context and connects policies, prior decisions, relevant data and procedural next steps into a coherent output tailored to the user’s role.
Recent research on enterprise agent design highlights that the primary bottleneck is not model capability but knowledge structure, which emphasizes the need to transform internal knowledge into structured, actionable formats that include governance constraints and organizational metadata.
Personalization, in this sense, is about making institutional knowledge usable without compromising control.
The implications for enterprise leadership are significant. Personalization is no longer a feature decision. It is an operating model decision.
MIT Sloan reported in March 2026 that 38% of surveyed companies have appointed a chief AI officer or equivalent role, yet there is little agreement on where that responsibility should sit within the organization. This lack of clarity often leads to fragmented AI strategies, where different functions build isolated solutions without shared governance.
Role-based AI requires coordination across IT, security, legal and business units. Without it, organizations risk inconsistency, duplication and exposure.
There is also a workforce dimension that cannot be ignored. As AI becomes more personalized, it also becomes more influential in how work is performed. A March 2026 analysis from the Thomson Reuters Foundation found that while AI adoption is accelerating across industries, only about one in ten companies report having policies in place to mitigate negative impacts on workers, and fewer than one-third publicly disclose dedicated AI governance resources.
This gap matters. Role-aware systems shape not just outputs, but decision-making patterns, escalation paths and access to information. They influence how authority is exercised within the organization.
For executives, the strategic question is not whether AI will become more personalized. It already is. The more important question is whether that personalization will be governed effectively. Generic AI systems can be tolerated because their limitations are visible. Role-based systems operate closer to decisions, which raises the stakes. They must be accurate, contextually appropriate and aligned with enterprise policies.
The path forward requires discipline. Organizations must define clear identity frameworks, align data access with role-based permissions and ensure that AI systems are auditable and explainable. Personalization should be treated as a controlled delivery mechanism for enterprise knowledge and action.
The next phase of enterprise AI will be defined by how well systems understand the organization they serve. AI that knows more is useful. AI that knows what matters, to whom and under what conditions becomes indispensable.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。