In partnership with
Enterprise AI · AI Spotlight
The Document Pile Just Got a New Boss. It Is AI.
Laserfiche just shipped AI agents that read your documents, understand what needs doing, and do it. No workflow builder required. No IT ticket needed. Just plain English.
Hey there, welcome back.
Every major software company is adding AI to their product. Most of the time it amounts to a chatbot bolted on the side. This week, Laserfiche did something more interesting.
They shipped AI agents directly into their content management platform. Not a demo. Not a beta feature for enterprise clients with six-figure contracts. Live, in production, available to Laserfiche Cloud users from May 7, 2026.
The idea is straightforward and worth paying close attention to: instead of you telling software what rules to follow, you tell it what outcome you want. The agent figures out the steps.
* * *
|
The Laserfiche AI Agent Snapshot
|
* * *
|
What Laserfiche Actually Built |
Laserfiche is an enterprise content management platform used by governments, hospitals, financial institutions, and mid-size businesses to store, organise, and process documents. Think: every form, contract, invoice, and employee record a large organisation touches. For decades, automating any part of that process required IT involvement, workflow designers, and considerable setup time.
The new AI agents change that equation. Users access them through Smart Chat, a chat interface already built into the Laserfiche repository. Instead of configuring rules, you describe what you want done. The agent analyses the documents, previews what it plans to do, waits for your approval, and then executes. All within the same security and compliance framework your organisation already has in place.
|
"The introduction of AI Agents to content management signals a change in how we handle the information lifecycle. We are moving beyond manual processes by offloading mundane work to agents that operate in a governance framework." Karl Chan, CEO, Laserfiche |
The key phrase in Chan's quote is "governance framework." This is not an AI tool that operates outside your organisation's rules. It inherits the permissions, access restrictions, and compliance controls of the user who activates it. If you cannot see a folder, neither can the agent you deploy.
* * *
|
What the Agents Actually Do in the Real World |
Laserfiche has been specific about three departments where the agents are already delivering value. These are worth unpacking because they illustrate the exact type of work AI is best suited to: high-volume, rule-based, and time-sensitive.
|
Legal Agents scan contracts and documents for internal inconsistencies before any human reviewer touches them. Flagged items are automatically routed for legal review. The result: lawyers spend time on judgment calls, not on reading 200-page contracts looking for clause mismatches. |
|
Accounts Payable Agents identify late or overdue invoices across the repository and automatically surface them to the correct team for resolution. No more manual invoice chasing. No more missed payment windows that damage supplier relationships or trigger penalties. |
|
Human Resources Agents scan employee records and identify specific data fields such as age, gender, and address. Based on that data, documents are automatically moved to the correct digital folder. Critically, the agent only acts within the security level of the user who deployed it, keeping sensitive records properly segregated. |
The common thread across all three: these are tasks that previously required either a person doing repetitive work or an IT team building and maintaining a custom workflow. The agent collapses both into a natural language instruction.
Talk to your AI tools the way you'd talk to a colleague.
You don't send a colleague a three-word brief. You explain the context, the constraints, what you've already tried. But typing all that into ChatGPT takes forever — so you don't.
Wispr Flow lets you speak your prompts instead. Talk through your thinking naturally and get clean, paste-ready text. No filler words. No cleanup. Just detailed prompts that actually get you useful answers on the first try.
Millions of users worldwide. Works system-wide on Mac, Windows, and iPhone.
|
The Detail That Changes Everything: Approval Before Action |
One design decision in this release stands out above everything else: the agents do not act without showing you what they plan to do first.
When you give a Laserfiche AI agent an instruction, it analyses the request, generates a detailed preview of every planned action, and requires your explicit approval before making a single change. This is not an afterthought. It is the architecture of the product.
In enterprise environments, the fear of AI is not that it will fail to do the work. It is that it will do the wrong work, at scale, irreversibly. The approval-first model directly answers that fear. You get the speed of automation with the oversight of a human decision at the critical moment.
|
* * *
|
What Comes Next and Why It Matters More |
The May 2026 release is explicitly described as the first phase. Laserfiche has outlined the roadmap clearly, and it escalates fast.
|
The Laserfiche AI Agent Roadmap
|
Phase 4 is the one to watch. An AI agent that runs in the background and acts when conditions are met is not a tool anymore. It is a colleague. One that never stops working, never misses a trigger condition, and never forgets to file something correctly.
|
"The 'where' of document storage is not going to be as important as it used to be. With automatically-extracted metadata, AI-assisted search and autonomous AI agents, you won't have to spend time organising data. You will be able to simply act on it." Justin Pava, Chief Product Evangelist, Laserfiche |
* * *
|
What Decision-Makers Need to Understand Right Now |
This release is not just a product update. It is a signal about where enterprise software is heading. If you manage a team, run a department, or make technology decisions for an organisation, here are the three things worth internalising.
|
1. The cost of automating work just dropped to near zero Before AI agents, automating a document workflow required a developer or a specialist to build it. With natural language agents, every task is a candidate for automation. The threshold for "worth automating" just collapsed. 2. Compliance is no longer a reason to avoid AI in regulated industries AI agents inherit the same permission structure your organisation already trusts. Government agencies, hospitals, and financial institutions that have been watching AI from a cautious distance have one fewer reason to delay. 3. The job of "organising information" is ending When an AI agent can route information reliably, at scale, with a full audit trail, the humans in that loop get redirected to decisions and work that actually requires judgment. |
* * *
The Bigger Picture
Laserfiche is not the most famous name in enterprise software. But this release is worth watching precisely because of that. When a focused, specialist platform ships agentic AI that is live, auditable, compliance-ready, and accessible to every user regardless of technical skill, it means the technology is no longer experimental.
The enterprise AI wave that analysts have been predicting for three years is not approaching anymore. It arrived on May 7, 2026, quietly, inside a document management platform that millions of government and business users already open every morning.
|
"AI agents did not replace the people doing this work. They replaced the hours of friction that prevented those people from doing more important work." |
* * *
|
🤖 Quick Question Is your organisation already using AI agents to automate document workflows? Hit reply and tell us. Are you piloting AI internally, watching from the sidelines, or still figuring out where to start? Every reply shapes the next issue. |
* * *
Until Next Time,
AI Spotlight
Practical AI. Real world impact. Every week.