The future of SaaS: why the era of AI agents is changing everything

The world of SaaS is changing. Slowly but radically. After years of growth based on traditional models collaborative software, reporting tools, analytics platforms, etc.the massive arrival of generative AI is disrupting expectations, interfaces, and uses.

But beware: this is not simply a matter of adding a conversational assistant in the corner of the screen. What is emerging is a paradigm shift. A new way of thinking about the product experience. A transition from SaaS to SaaS + AI Agents.

From tool software to assistant software

For a long time, enterprise software was a toolbox. It was used to build, capture, organize, and analyze. It replaced Excel spreadsheets, emails, and Word documents.

But it still required a lot of manual effort. Users had to know the methods, formulas, tables, etc. The software was there to centralize. Not to help.

With generative AI, we can take it a step further. We can move from a data entry mindset to an assistance mindset. We can ask the software to “suggest an initial version,” “clean up this data,” or “summarize what you've understood.” And we get a contextualized, clear, and directly usable response.

The AI Agent, a new work companion

But that's the challenge: for this to work, AI needs to be more than a generative engine. It needs to become an agent. A specialized assistant who knows your context, your business, and your data structure. An assistant with a specific mission, a clear framework, and a well-defined role in your workflow.

Less interface, more interaction

When AI becomes an agent, everything changes. You no longer click to search for a feature. You dialogue. You guide. You review. The interface disappears in favor of interaction.

This doesn't mean that the software disappears. But it becomes more fluid, more adaptive, and more intelligent. It understands the context, anticipates needs, and suggests shortcuts. It works with you, not just for you.

In a tool like Albert, this means, for example:

• You import raw job data, and the Job Archi Extractor assistant suggests a hierarchical structure.
• You start a project without a job description, and the Job Builder assistant generates a draft aligned with your strategy.
• You finalize your plan, and the SWP Report Generator assistant produces a clear, distributable report that is ready to share.
  

Fewer clicks, more impact. That's the real gain.

The SaaS of tomorrow will be modular, assisted, and invisible

This movement can be summarized in three trends:

• Modular: each assistant responds to a specific need. It can be called upon individually or in a series. Users no longer follow a fixed path.
• Assisted: the software suggests, reformulates, and optimizes. It doesn't wait for users to know everything. It guides them.
• Invisible: the best software is the software you don't notice. It allows you to focus on the decision, not the interface.
  

AI agents embody this future. They shift the value of software to its use. And they give power back to users without complicating their daily lives.

So what about Albert? Assistants or agents?

At this point, a clarification is in order. At Albert, we have indeed developed powerful AI assistants capable of reading files, generating new ones, structuring, reformulating, synthesizing, and more. Their impact is real and their role is well defined in HR workflows.

But can we really call them “agents”? Not quite.

The line is certainly fine, but essential: our assistants do not (yet) have autonomy in the strict sense of the word. They are not inextricably linked to the tool with independent decision-making or triggering capabilities. This choice is intentional, for reasons of legality, security, reliability, and business control.

So no, our assistants are not agents in the strict sense of the word. But they are part of this agent-based dynamic. They are specialized, contextualized assistants embedded in mission-driven logic. They reduce clicks, increase impact, and streamline decisions. They don't take control, but they provide a very useful helping hand.

What this means for HR teams

In concrete terms, for HR, the emergence of AI agents means:

• Less time wasted looking for the right formula, the right format, the right method.
• Greater clarity in deliverables and consistency between steps.
• Greater recognition of the value of HR work, as results are shareable, understandable, and actionable.
  

This does not replace the HR function. It reinforces it. It allows teams to refocus on their mission: understanding skills, anticipating needs, and supporting managers.

Conclusion: AI agents, the new backbone of modern SaaS?

We no longer design tools the way we used to. Users want results, not settings. They want meaning, not clicks. They want a seamless experience, not a 20-page tutorial.

AI agents, when well designed, well integrated, well supervised, or simply conceived as specialized intelligent assistants, can fulfill this role. At Albert, we have opted for a hybrid model that is powerful, pragmatic, and secure. Because it's useful. Because it's effective. Because it's simply the logical next step for HR software.

‍