Editorial
AI agents — autonomous software systems that can perceive, reason and act — are becoming an integral part of the workforce, a new talent pool alongside human workers. But the way we organize and deploy them is about to evolve. AI agents have the potential to fundamentally change how we access and utilize data across organizational boundaries to get work done.
These agents are building blocks that can be put together in infinite ways to execute a workflow when given access to the right data. The potential to increase productivity and break down silos that have constrained traditional workflows and technology for decades is significant. But realizing it requires a change in how we think about work, organizational boundaries and data access.
Today's Reality: AI Agents in Traditional Hierarchies
AI agents are already becoming an integrated part of the workforce. Organizations are largely using AI agents to augment tasks that people have historically executed, which is why many are placing agents within existing team structures and reporting lines. This approach provides immediate benefits like transparency, accountability and governance. AI agents get access to the same data, processes and workflows as the role they are augmenting — nice and easy.
While this traditional approach offers a comfortable starting point, it also presents inherent limitations. Placing agents within existing org charts constrains the data they can access. Employees today are familiar with these access limitations, which is why seemingly simple cross-functional projects can result in a hornets nest of complexity. Protected data and legacy processes force workers to jump through hoops and navigate scattered systems to access the information they need. These data dependencies have shaped organizational structures for decades, creating boundaries that AI agents have the potential to transcend.
Functionality like zero copy and data fabric allows data to be used in a process or workflow without exposing it to any human. AI can access customer data from sales systems, financial information from accounting platforms and operational metrics from service databases — all within a single workflow, without the traditional handoffs and delays.
Related Article: Is Your Data Good Enough to Power AI Agents?
The Building Blocks Approach
Think of these agents not as "employees" but as Lego bricks that execute specific tasks while bridging data and systems. They will be combined in countless different ways. They don’t need to be constrained by existing organizational boundaries, and eventually, they will become extremely dynamic. The more data agents have access to, the more dynamic they can be. Agents will handle individual tasks or parts of workflows — for example, an agent that fetches data or sends emails represents a foundational building block of this new world. Instead of separate agents repeating the same task for each dataset or technology, one agent can perform it using any data in the company.
This represents a fundamental shift from how we've historically approached work design. In the past, job descriptions and organizational structures were often shaped more by what data or systems someone could access than how they created value for the organization. AI agents eliminate these artificial constraints, allowing us to design workflows based on optimal outcomes rather than systems or trying to work around data limitations.
For example, the hiring process has traditionally been executed by different personas, each with access to unique data or skills (e.g., hiring manager, recruiter and sourcer). The recruiter has access to compensation data, while the sourcer has access to candidate pool records. In the future, this process will consist of task-based agents orchestrated to complete the entire process: one agent scans candidate profiles, while another writes personalized outreach emails (which may be customized by role or function leveraging data from multiple systems), a third schedules interviews and a fourth compiles candidate feedback. Each has a specific task and uses all available data to execute it effectively. Humans are there to build relationships, make hiring decisions and ensure agents are delivering high-quality work.
The orchestrator coordinates these specialized agents to execute the workflow, data and systems efficiently, saving time and reducing human coordination needs. Agents of the future will be combined dynamically from these task-based building blocks into meaningful multi-agent workflows, learning and adapting with context-specific flexibility based on comprehensive organizational data and human oversight. Think about how the experience could improve by addressing the entire problem holistically, rather than approaching it piecemeal — by system or with limited data.
Managing the AI Ecosystem
You can't do this without the right technology and governance that manages AI agents. Organizations need an AI Control Tower that centralizes strategy, governance, performance and management across the entire AI ecosystem while driving enterprise-grade compliance and accountability. Organizations will need centralized technology platforms that provide:
- Enterprise-wide visibility: The ability to monitor and manage all AI agents, models and workflows across the organization with consistent policies.
- Governance and compliance frameworks: Proactive risk management capabilities that ensure security, privacy and regulatory compliance across the entire AI lifecycle.
- Lifecycle management tools: Systems that support the full journey from ideation to deployment to optimization, enabling contextual decision-making and enforcing appropriate guardrails.
- Performance analytics: Real-time operational insights that measure AI performance against key business outcomes like productivity and revenue impact.
- Strategic alignment mechanisms: Tools that help organizations ensure their AI initiatives align with broader business objectives and strategy.
Without integrated management capabilities and enablement, organizations will struggle to effectively deploy and govern their increasingly complex AI workforce components.
Webinar
Oct
16
Agentic AI Playbook: Real-World Customer Service Use Cases You Can Deploy Now
Boost self-service by 30% and slash call volume by 63% with agentic AI.
Register
Webinar
Oct
22
Beyond Storage: Smarter Content, Bigger Impact with DAM + AI
Discover how the DAM + AI duo makes content smarter, stronger and more accessible.
Register
Webinar
Nov
6
Fix the Content Bottleneck: Build a Better WebOps Strategy
Content stalled? Dev overloaded? You’re not the only one. Learn how streamlined WebOps bridges the publishing gap.
Register
Webinar
On demand
CMS Briefing: A Live Look at What’s Next in AI-Driven Platforms
Learn how leading organizations are using AI‑driven tools to publish faster, personalize smarter and stay secure.
Watch Now
Webinar
On demand
Ready or Not: How Data-First Organizations Are Unlocking Agentforce Potential
Learn how to cut through the noise, activate Agentforce and build a Salesforce AI strategy that actually delivers.
Watch Now
Webinar
On demand
AI in Customer Service: Faster Resolutions, Happier Customers
Don’t let rising demand burn out your team. See how to build a smarter, more resilient support org.
Watch Now
Related Article: How to Build Multi-Agent Workflows That Don't Fall Apart
Reimagining the Organization
My team has done the analysis and found that 0% of positions at our company are untouched by AI, meaning AI will augment every single role. Integrating AI agents isn't a nice-to-have. The organizations that will thrive are those that are thoughtful in integrating AI agents into their workforce, but not by simply plugging these agents into existing structures as direct substitutes for human workers, but by fundamentally reimagining how data flows and how cross-functional collaboration works.
As we move from today's approach of fitting AI agents into familiar hierarchies toward tomorrow's vision of cross-functional, data-integrated agents working in dynamic combinations, organizations must be willing to fundamentally rethink how work gets done and how data flows.
Data security and privacy will be a challenge — but one I believe we can balance. Organizations or functions that hoard their data will prevent the promise of AI and ultimately negatively impact business outcomes. Success will come not from digitizing yesterday's organizational models, but from pioneering new ways for humans and AI agents to collaborate in workflows designed for this boundary-free reality.
Learn how you can join our contributor community.
