The shift toward an AI-driven workplace is happening faster than you might expect. AI agents—software programs that take actions, make decisions, and trigger workflows without human involvement—are multiplying at a staggering rate. Gartner projects that the average Fortune 500 enterprise will operate more than 150,000 AI agents by 2028, up from roughly 15 in 2025.
But here’s the catch: only 13 percent of organizations believe they have the controls in place to manage this rapid agent proliferation. That gap is exactly what the Second Edition of Insygna’s white paper on Agentic workforce management aims to address, offering you a practical roadmap for enterprise AI governance and workforce readiness.
The Hidden Costs and Compliance Risks of Unmanaged AI Agents
Leaving AI agents untracked is a gamble that few organizations can afford. Without proper oversight, these autonomous tools can quietly drain budgets, expose sensitive data, and create serious compliance exposure. The appeal of letting agents run independently often masks a costly reality.
One of the most immediate dangers is AI agent cost overruns. Each autonomous decision an agent makes may trigger compute usage or API calls that you never authorized. Without monitoring, these small actions accumulate into significant operational expenses that hit your bottom line without warning. You might only notice when the monthly cloud bill arrives — and by then, the damage is done.
Beyond financial leaks, there is the risk of data leakage AI agents. An unmanaged agent can access internal databases, customer records, or proprietary systems without any governance guardrails. If it makes a decision that exposes that information to the wrong channel or stores it in an unsecured location, your organization faces a breach that is both costly and reputationally damaging.
Perhaps the most concerning issue is the growing AI compliance risk. As agents make autonomous decisions, they can inadvertently violate regulations such as GDPR or the upcoming EU Digital Omnibus. Without an autonomous agent audit trail, you have no way to trace which decision caused the violation or who — or what — is responsible. This makes effective agentic workforce management not just a technical preference but a core business necessity.
Who Owns the AI Agent? Shifting Governance from IT to Business Functions
Once an AI agent starts working, the real management challenge moves from the IT team to the business owners. This is a significant shift in how organizations think about agentic workforce management. The white paper directly asks: “Who owns the agents once they go live?” Its conclusion is clear — IT builds and deploys the technology, but managing its day-to-day performance and impact is a business responsibility. You need to understand this distinction to avoid confusion and inefficiency.
To make this work, companies are rethinking leadership structures. For example, Atlassian, Moderna, and ServiceNow have expanded the role of their top HR executive to lead AI workforce strategy. This move gives AI agent ownership a clear home within the organization, ensuring that the people who understand your workforce and culture are steering the technology. It signals that business ownership of AI is not just a technical checkbox — it is a strategic priority.
Assigning clear ownership prevents agents from becoming “orphan bots” with no accountability. Without a designated manager, an agent can drift off course, make unchecked decisions, or simply stop being useful. By embedding agentic governance responsibility into your business functions, you create a system where every agent has a champion who monitors its behavior, updates its rules, and ensures it aligns with your goals. This practical step is essential for any organization serious about scaling an AI workforce strategy HR teams can support effectively.
Preparing for the EU Digital Omnibus on AI: What Companies Need to Know
The European Union’s new omnibus regulation will impose concrete requirements on AI agent management, and companies must act now. The Second Edition of Insygna’s white paper reflects the European Union’s June 2026 adoption of the Digital Omnibus on AI. This regulation marks a significant shift, moving from general AI guidance to specific, enforceable rules for autonomous software agents. For any organization using AI agents at scale, understanding the EU Digital Omnibus AI requirements is no longer optional — it’s a compliance necessity.
The regulation mandates three core pillars for managing autonomous agents: transparency, human oversight, and risk management. First, you must be able to clearly explain what each agent does, how it makes decisions, and which data it accesses. Second, a human must remain in the loop for critical actions, meaning your workflows need clear escalation points. Third, you need documented risk assessments for every agent in your fleet. These are not abstract ideals; they are practical obligations tied to the agentic workforce management practices you build today.
To get ahead of the AI agent regulation EU, start by auditing your current agent fleet. Catalog every autonomous agent in production, note its purpose, and evaluate its potential for error or bias. Then, implement governance controls that track agent decisions and allow for manual overrides. Many companies find that tools for logging agent activity and setting permission boundaries become essential here. By taking these steps before the June 2026 deadline, you transform AI compliance 2026 from a hurdle into a structured process. The focus keyword here is proactive preparation: your goal is to meet the autonomous agent transparency requirements while keeping your operations efficient.
Leveraging HR, Finance, and Procurement Processes for AI Agent Governance
Now, how do you actually govern this new digital workforce without reinventing the wheel? The answer may be closer than you think: your existing enterprise processes. You already have systems for managing people, budgets, and vendor relationships — and those same frameworks can be adapted for agentic workforce management. Think of AI agents as a new class of digital workers that need their own version of onboarding, oversight, and cost controls.
Start with HR processes. Onboarding a human employee involves training, role assignment, and performance reviews. You can apply a similar cycle to your AI agents: define their purpose, set permissions during a “digital onboarding,” run regular checks on their decision quality, and decommission agents that underperform. Companies like Atlassian, Moderna, and ServiceNow have already expanded the role of their top HR executive to lead AI workforce strategy — a clear sign that HR for AI agents is becoming a core function.
Finance controls are equally important. Without proper tracking, agent usage can spiral into unexpected costs. By applying budgeting, cost allocation, and chargebacks — the same tools you use for software subscriptions — you can monitor each agent’s compute consumption and set spending limits. This makes finance AI agent cost allocation a straightforward extension of your existing expense management.
Finally, treat your agent subscriptions like any other vendor relationship. Procurement lifecycle management can handle the selection, contract terms, and renewal of agent-as-a-service offerings. Just as you’d audit a SaaS tool, you can review an agent’s compliance with your policies. This procurement AI agent subscription approach keeps your digital workforce management processes consistent, scalable, and audit-ready.
The Six Pillars of Agentic Maturity: A Self-Assessment for Organizations
Beyond the procurement audit approach, you need a broader framework to measure how ready your organization really is for a digital workforce. That’s where Insygna’s self-assessment tool comes in. The paper includes a maturity self-assessment that scores your company across six areas of agent governance. Think of it as a diagnostic check for your agentic workforce management strategy — it helps you see where you stand today and where you need to focus next.
On a similar note, Researchers Turn Old Junk Drawer Phones Into Cloud Platform explores this topic with concrete examples.
Although the six specific pillars are detailed in the white paper, they span categories that matter most when managing AI agents at scale: compliance, cost control, security, and lifecycle management. These aren’t isolated concerns. For example, a gap in security governance can directly affect your compliance posture, while weak lifecycle management can inflate costs. The AI agent maturity model inside the paper ties these together so you get a complete picture of your digital workforce readiness score.
How do you use it? You score your organization on each pillar. The results highlight your strengths and, more importantly, uncover gaps you might have missed. Maybe your compliance processes are solid, but your cost controls are loose. Or perhaps your security measures are strong, but you lack a clear lifecycle management policy. The agent governance assessment gives you a structured way to spot those imbalances.
Once you have your scores, you can prioritize actions. Don’t try to fix everything at once. Focus on the pillars that need the most attention first. Then, as you improve, retake the assessment to track your progress. This cycle of measuring, acting, and measuring again keeps your six pillars AI management approach practical and grounded. The goal isn’t perfection on day one — it’s steady, measurable improvement over time.
Insygna’s Platform and the Path to Agentic Workforce Control
Putting those iterative principles into practice requires a tool built for the job. Insygna has developed an award-winning platform that turns the white paper’s governance framework into a daily working reality. Instead of hoping your six-pillar approach stays on track, the Insygna AI agent platform gives you a central dashboard for agent discovery, policy enforcement, cost monitoring, and compliance reporting. It translates broad guidelines — like visibility or accountability — into concrete, enforceable rules across every AI agent your team uses.
The platform’s capabilities have already earned industry recognition. Insygna won the HR Tech Europe 2026 Startup Competition, and it has been shortlisted for a 2026 HR Tech Product of the Year award. These nods confirm that agent governance software is no longer a niche concept — it’s becoming an essential part of modern workforce management.
As an AI workforce management tool, the platform connects directly to the measurement and adjustment cycle described earlier. You can set governance policies, view real-time cost data, and run compliance reports without digging through scattered logs. That visibility supports the kind of steady, data-driven refinement that keeps your agentic workforce management practical and effective over the long haul.
If you want to see the solution in action, Insygna will be exhibiting at TechCrunch Disrupt and HR Tech in October 2026. Both events offer a chance to explore how the platform brings that six-pillar logic to life — and how it helps you move from abstract governance to real-time, measurable control over your AI workforce.
Frequently Asked Questions
How can my organization prepare for the projected surge in AI agents by 2028?
Start by mapping where AI agents could add value in your workflows, then assign clear ownership for each agent’s lifecycle. Build a governance framework that includes identity management, access controls, and audit trails from day one. Running small pilot programs helps you learn how to scale agentic workforce management before the volume grows.
How can existing HR, Finance, and Procurement processes be applied to AI agent governance?
You can adapt familiar workflows: HR-style onboarding and offboarding procedures work well for registering and decommissioning agents. Finance processes like budget approvals and cost allocation help track agent resource consumption. Procurement’s vendor management approach translates directly into overseeing third-party AI agent services and their compliance.
What are the biggest risks of leaving AI agents unmanaged?
Unmanaged AI agents can take actions that bypass your security policies, create data leaks, or generate unauthorized costs. They may also produce inconsistent outputs that damage your brand reputation or violate regulations. Without proper agentic workforce management, you lose visibility into what each agent is doing and how to correct it quickly.
