Your AI Workforce Is Already Here. Can You See It?
October 2, 2026 · Keith Pasley
Your AI Workforce Is Already Here. Can You See It?
Most companies are planning for an AI workforce as if it is something coming in the future.
I think it’s already here.
Employees are using ChatGPT, Claude, Copilot and dozens of other AI services.
Developers are connecting coding agents to repositories and development tools.
Business teams are experimenting with agents.
MCP servers are giving AI access to enterprise tools and data.
And some of this is happening without security or governance teams knowing it exists.
That changes the AI governance problem.
The first question isn’t:
How do we govern our approved AI platforms?
It’s:
Do we actually know how AI is being used across the organization?
F5’s AI Workforce Security capabilities are interesting because they approach the problem from that direction.
F5 says its AI Security Platform can continuously discover sanctioned and shadow AI usage, monitor MCP servers and agents, and classify AI workflows based on intent. It can then apply controls to sensitive information and agent-driven tool use. (F5, Inc.)
The intent piece matters.
Knowing that someone accessed an AI service tells you something.
Knowing what they were trying to accomplish with it tells you considerably more.
An employee asking an AI model to rewrite a public marketing paragraph creates a very different risk than someone asking the same model to analyze confidential customer records.
The application might be identical.
The risk isn’t.
Agents make this even more important.
An agent can access an approved model and still attempt an inappropriate action with an approved tool.
That means enterprises increasingly need visibility across several dimensions:
Who is using AI?
Which models and agents are being used?
What data are they accessing?
What are they trying to accomplish?
Which tools are agents invoking?
And what should actually be allowed?
This is where AI governance has to move beyond maintaining a list of approved tools.
Governance needs to become operational.
Discover the AI.
Understand the intent.
Classify the risk.
Apply policy.
Enforce the decision.
Monitor what happens next.
F5’s approach is another signal that AI security is expanding beyond protecting the model itself.
The security boundary now includes the people, agents, models, data, APIs, MCP tools and applications participating in the workflow. (F5, Inc.)
That is a much larger architecture.
And it leads to a very practical enterprise AI question:
Before you build a strategy for governing your future AI workforce, can you see the AI workforce already operating inside your organization?
Technology does not create business value. Better decisions do.
Source: F5, F5 AI Security Platform — AI Workforce Security. F5 describes its capabilities including AI workflow intent classification, continuous discovery of sanctioned and shadow AI, monitoring of MCP servers and enforceable controls over sensitive content and agent-driven tool use.
https://www.f5.com/products/ai-security-platform?trk=public_post_comment-text&utm_source=chatgpt.com