Breaking news from Glean

We talk a lot about AI agents in this industry. Most of the time, what we see are smarter chatbots that still rely on humans to stitch work together across tools.

This morning, my partner Glean announced something that feels like a real step forward.

They introduced what they call the industry’s first autonomous agents with full enterprise context. In simple terms, these are AI systems that understand how your company actually works and then get real work done across your existing tools.

This is not just a new feature. It is a different way of thinking about how work flows through an organization.

What exactly did Glean announce

Here is the short version.

1. Autonomous agents with full enterprise context

Glean’s new agents can:

• Interpret instructions in plain language

• Decide which tools to use

• Take actions across systems like Salesforce, Jira, Confluence, GitHub and more

• Explain what they did and why

You can spin these agents up in an auto mode inside Glean’s agent builder. So you are not writing scripts. You describe the outcome you want and the agent figures out the steps using more than 100 built in actions.

2. Glean Enterprise Context as the engine

All of this runs on top of Glean Enterprise Context.

Think of it as a live map of how your organization operates:

• Your data across tools

• Your people and how they work together

• Your processes and workflows

As agents complete tasks, they build memory that is unique to your business. Over time, the system learns how things get done in your company and becomes more accurate and more useful.

3. Connectors that reach into core systems

Glean is also expanding its native integrations.

Along with existing productivity tools, they now support Microsoft Dynamics 365, Netsuite, Ironclad, Affinity, Procore and Canva.

In practice, this means an agent is not stuck in a single department. It can follow work across sales, finance, legal, design and operations.

4. Tool Search to choose the right action

One of the hard problems with agents is choosing the right tool for each task.

Glean’s new Tool Search feature lets agents automatically:

• Discover which apps are relevant

• Select the right actions

• Use specialized searches like calendar, code and expert search

This matters if you want a single agent to navigate hundreds of tools and actions without hand holding.

Guardrails that make this enterprise ready

A lot of AI demos look great on stage and fall apart when they meet real security and governance requirements.

Glean has put a lot of thought into this part.

Agent alignment and controls

• Every action is checked against the agent’s original scope

• Out of scope behavior and risky steps can be blocked

• Admins can define who is allowed to create agents and how they are used

This is important if you want teams experimenting with automation without losing control.

Protection of sensitive data

Glean continues to enforce user level permissions and now supports sensitivity labels in Google Drive and Microsoft Purview, which extend to OneDrive and SharePoint.

On top of that, they have dynamic detection for sensitive content. So the same protections that apply to your data today follow the agent as it works.

Why this matters for leaders

If you are a CIO, CDO, Head of Data or an operations leader, here is why I think this announcement is important.

1. It targets the real bottleneck

Most teams are not held back by one big system. They are held back by the small steps between systems.

Someone has to:

• Pull data from one tool

• Cross check it in another

• Update a third system

• Then notify the right people

Glean’s agents are designed to own these multi step workflows. That is where real productivity shows up.

2. It cuts across functions, not just one use case

Because of the breadth of connectors and actions, this is not a point solution.

You can imagine:

• Sales operations cleaning up records and routing accounts

• Finance teams automating parts of reconciliation and reporting

• Legal teams moving contracts through standard steps

• Engineering and support teams syncing issues between tools
  

All running on the same agent platform, with the same governance model.

3. It moves you along the AI maturity curve

Many enterprises are stuck in assistant mode.

You have a chatbot that answers questions. Maybe it drafts content. But it does not own outcomes.

This Glean announcement is about moving from:

• “AI that helps you think”
  to

• “AI that can own well defined pieces of work”

With context, memory and guardrails built in.

My take

We are going to look back at this phase and see a clear shift.

First, we had AI as a better search box.
Then AI as an assistant inside individual tools.
Now we are starting to see AI as an autonomous teammate that sits on top of your systems and keeps work moving.

Glean’s autonomous agents with Enterprise Context fit squarely in that third category.

If you are exploring agents in your own organization, this is the kind of architecture to study: deep context, broad connectors, careful alignment and strong data protections.

This is how AI moves from a side project to part of the fabric of how your business operates.