Is this the end of SaaS as we know it?

Picture this...

It's 1876, and Alexander Graham Bell just invented the telephone. Telegram companies are probably in emergency meetings, convinced this newfangled device is just a passing fad. Spoiler alert: It wasn't. Fast forward to today, and we're witnessing a similarly significant moment in the evolution of the software industry.

Over the past twenty or so years, Software-as-a-Service (SaaS) has been a defining pillar of the tech industry, ballooning into a market delivering solutions that organisations, big and small, now depend on. Yet there’s an argument steadily gaining momentum: that generative AI could rapidly erode the very underpinnings of SaaS, ultimately transforming, or even deconstructing, an industry once considered unassailable.

Below is a look at how AI agents are changing the status quo, what this could mean for SaaS vendors, and why some may still emerge as major aggregators in a world increasingly shaped by agent-to-agent communication. We also reflect on the opportunities (and challenges) for those organisations looking to make the switch from SaaS to AI, just like those that made the switch from telegram to telephone all those years ago!

The problem with fragmentation: Why some organisations are re-evaluating SaaS

As AI adoption accelerates, a recurring challenge has become evident in many organisations: data fragmentation. Over time, companies have accumulated sprawling ecosystems of SaaS tools, each managing different aspects of their operations. For example:

• Customer interactions might live in CRM platforms.
• Project workflows are scattered across task management tools.
• Financial data is buried in ERP systems.
• Internal knowledge exists across wikis, document storage, and various chat logs in places like Slack.

On paper, SaaS promised to streamline operations. In reality, it has often siloed critical information, duplicated records and introduced some complexity. As businesses look to harness AI, they’re realising that dispersed, inconsistent, and redundant data doesn’t power great AI outcomes. Feeding AI an unstructured mess of conflicting records from different platforms could well lead to unreliable insights.

A growing number of organisations are re-evaluating their software stack, reducing reliance on multiple SaaS tools, and focusing on unifying their data into a more structured, accessible framework. The goal? Consolidate knowledge, reduce friction and create a foundation that AI can interact with seamlessly.

This doesn’t mean SaaS is disappearing entirely, but its role is changing. Some software providers will struggle to remain relevant, while others will transform into AI-ready platforms that aggregate and standardise enterprise knowledge. The difference between survival and obsolescence may come down to which ones embrace this shift.

AI-augmented workflows

What’s happening:

Today, AI is currently being introduced as an enhancement to existing SaaS solutions, often referred to as copilot functionality. This stage sees AI embedded into established software to expedite repetitive tasks, assist with decision-making and improve overall productivity. The result is reduced friction for end users and incremental efficiency gains for organisations.

Why it matters:

• Operational efficiency: By offloading routine or high-volume tasks to AI, staff can focus on other priorities.
• Rapid adoption: Because these AI tools are layered on top of familiar systems that staff use day in and day out, change management is less disruptive.
• Enhanced user experience: Advanced analytics, natural language queries and predictive insights make it easier to derive value from SaaS products that organisations feel heavily invested in.

From our perspective here at Storm ID, organisations should view AI-augmented workflows as an entry point to fundamental modernisation of their processes. Early adopters can test feasibility, measure ROI and develop internal capabilities in rapid preparation for more autonomous AI deployments.

Autonomous AI operations

What’s happening:

For organisation’s that get it, AI’s role will evolve from guided assistant to autonomous AI agent. Rather than simply aiding human decision-makers, AI agents will be able to carry out entire tasks or business processes on their own. For instance, instead of manually navigating multiple dashboards in a SaaS solution, users will issue instructions and the AI will handle the execution behind the scenes.

Why it matters:

• Task automation: AI will increasingly manage end-to-end workflows, reducing the need for dedicated human operators.
• Unbundling of expertise: The domain-specific knowledge once ingrained in user interfaces and human actions will shift into AI-driven processes.
• API-Centric ecosystems: Integrations and data sharing become critical as AI agents interact with back-end APIs directly, minimising traditional front-end usage.

From our perspective here at Storm ID, this transition highlights the need for robust data pipelines, well-structured APIs and clear governance frameworks. Large aggregator SaaS platforms that can expose secure, seamless APIs for AI interactions may become even more valuable, as they’ll function like data hubs in an increasingly AI-driven ecosystem.

Interface-neutral intelligence

What’s happening:

In this step, software interfaces themselves may become less visible, or simply unnecessary, to most end users. As AI agents work directly with back-end systems, traditional dashboards and user interfaces that are synonymous with SaaS solutions, may no longer be the primary touchpoints. Instead, AI autonomously orchestrates data processing, analytics and operational decision making.

Why it matters:

• Invisible software: Humans receive insights or results, but the AI handles the underlying navigation, data calls and manipulation.
• Reduced eeliance on classic UX: The business value in software shifts away from front-end design towards secure, scalable and accessible data infrastructure.
• Strategic imperative: Organisations that adapt to this interface-neutral model can rapidly deliver tailored solutions, removing barriers between data, analytics and action.

This stage represents a significant pivot to getting architecture, security and interoperability right. While user interfaces won’t vanish entirely, especially in highly regulated domains or specialised use cases, the ability to leverage AI behind the scenes will be a key differentiator in a post-SaaS or evolving-SaaS world.

But SaaS isn’t going away completely…is It?

Plenty of dissenting voices say we’re just witnessing SaaS evolve, rather than disintegrate. True, the fundamental need for secure, cloud-based data systems will not vanish. Compliance, permission controls and robust infrastructure will remain essential. In fields like finance or healthcare, where regulation is strict, there will still be reasons for governed user interfaces and specialised workflows.

Yet, the real threat isn’t to the entire concept of software delivered via the cloud so much as the traditional SaaS business model and need for specialised interfaces. The assumption that knowledge workers will individually log in and operate the software themselves is unravelling. AI agents can:

• Use existing software autonomously on a per-task or outcome basis.
• Bypass SaaS UIs and communicate directly with underlying databases or APIs.
• Rapidly spin up customised, internal solutions via AI-driven development, shortening the distance between idea and production.

Where do we go from here?

1. Agent-friendly APIs

The winners will be those who make it easy for AI agents to tap in and leverage their data sources. If platforms don’t offer robust API access, and if they don’t handle intricacies like security and data governance, customers’ AI agents will find (or build) alternatives.

2. Trust and data

Incumbent SaaS providers still have huge data troves that AI loves to consume. If they can adapt by offering agent-ready data models, governance and reliability, they can remain relevant, perhaps even indispensable.

3. Specialisation

There will always be specialised verticals (regulated industries and niche use cases) where domain expertise truly matters. Still, even these will likely be restructured around AI. Expect to see domain-specific AI, custom language models and specialised prompts driving adoption, not conventional user interfaces.

The future: SaaS, but not as we know it

It’s not that SaaS is going the way of the telegram tomorrow. But the user interface-centric model, in which humans click around in dashboards to unlock insights and drive value, will lose ground. As AI agents handle tasks directly, the shape of SaaS will morph.

At the same time, larger SaaS operators have an opportunity. Those that unify organisational knowledge (eliminating data fragmentation, integrating AI-ready APIs and offering structured intelligence) could thrive in this new era.

In short, we’re not just facing an incremental tweak; we’re hurtling toward a shift in how software is built, sold and consumed. SaaS companies that pivot swiftly may remain relevant. Those that don’t may find themselves trying to peddle features to an AI that doesn’t really need them.

We’re at the beginning of something fundamentally different. SaaS isn’t dead (yet!), but it is evolving. The real winners will be those who adapt not just to AI, but to the new way AI interacts with enterprise knowledge.

At Storm ID, we help organisations harness digital, data and AI to stay ahead. As AI agents and automation reshape workflows, we’re enabling businesses to integrate AI-ready infrastructures, streamline operations and unlock new efficiencies. Learn more about our Agentic Workflow Engine and Dynamic Forms solutions.

Get in touch to explore how we can support your AI transformation.