A handful of friends and colleagues recently shared the same headline: Klarna is supposedly unsubscribing SaaS for AI. The buzz focuses on how AI—especially generative AI—has automated key operations and replaced some enterprise tools. This raises an obvious question: is AI about to make SaaS obsolete?
At first glance, it’s a tempting narrative. But if we look at how major technological changes have played out historically, we often see a more gradual shift—new solutions tend to layer on top of existing ones rather than instantly wiping them out. Still, it’s worth noting that generative AI represents a different kind of leap, one that can rewrite code, orchestrate across systems, and automate knowledge-based tasks in ways we haven’t seen before. Historical patterns aren’t guaranteed to repeat exactly, and there’s a chance AI’s impact will be faster or more disruptive than earlier shifts.
Looking Back to Look Ahead
When cloud computing first gained traction, many predicted it would quickly replace all on-prem systems. In reality, plenty of enterprises rely on a hybrid approach, mixing legacy on-prem setups with newer cloud-based solutions. Similarly, mobile devices were touted by some as the end of personal computing, yet laptops and desktops remain in heavy use. Time and again, these supposed “endings” turn out to be more like ongoing evolutions.
This pattern hints that AI might not wipe out SaaS overnight. Instead, it could reshape the landscape, pushing businesses to rethink how they integrate and use software—but not necessarily tossing out everything that came before. That said, it’s crucial to recognize that AI’s ability to learn, generate, and automate at scale could mean more rapid change than historical analogies suggest. We’ve never before seen a technology that can effectively recreate software logic on the fly, so it may challenge the typical slow pace of enterprise transitions.
Ecosystems and Gradual Evolution
A helpful way to see this is by thinking of SaaS platforms as deeply ingrained components of a company’s infrastructure. They store records, handle complex workflows, and comply with industry regulations. Simply replacing them is rarely practical, especially when there are significant compliance and operational details at stake.
Does that mean it’ll all stay the same? Not at all. AI can automate tasks across systems and potentially reduce the need for some specialized interfaces. It might even change how we think about user interaction—since an AI could handle data flow behind the scenes. But that doesn’t automatically remove the value of domain expertise, industry-specific workflows, or decades of refinements baked into many SaaS platforms.
Where AI Might Challenge SaaS
AI can orchestrate routine tasks, generate content, and serve as a bridge across multiple systems. It’s easy to envision a company-wide assistant that connects HR, finance, and CRM tools—reducing manual effort and potentially becoming the primary “front end” for everyday enterprise interactions. In practice, though, these capabilities still require careful integration, solid data management, and oversight to work reliably in complex environments. Even if AI takes center stage on the user-facing side, it doesn’t eliminate the underlying role of SaaS: the deep business logic, structured workflows, and compliance safeguards that remain essential for real-world operations.
SaaS platforms are more than just user interfaces. They’re shaped by years of domain expertise, proven security measures, and industry-specific compliance frameworks. While AI can learn patterns over time, it typically doesn’t arrive pre-loaded with the legal, regulatory, or specialized knowledge that SaaS vendors have spent years perfecting. Ultimately, AI may streamline how we interact with enterprise software, but it still depends on the robust foundations that SaaS provides behind the scenes.
The Hybrid Future
Rather than a scenario where AI simply “replaces” SaaS, it seems more likely the two will converge. Many SaaS platforms are already integrating AI to offer improved analytics, suggestions, and automated workflows, including us at Mekari. Meanwhile, AI systems will continue to rely on SaaS data stores, APIs, and built-in logic to carry out complex tasks.
This approach benefits everyone. AI can streamline operations, but it remains anchored in the proven reliability of SaaS. Meanwhile, SaaS platforms that embrace AI can deliver more automated, data-driven capabilities to their customers.
Lessons from the Past (and Present)
We’ve seen again and again that new technologies rarely sweep away older ones in one dramatic move. On-prem systems coexist with the cloud, and PCs still coexist with mobile devices. Still, with AI, there is a case to be made that the next leap could be quicker or more sweeping, especially if generative AI manages to absorb compliance logic, regulatory guidelines, and business processes at scale. That possibility shouldn’t be ignored, since it may compress the timeline for meaningful change.
Yes, there will be disruption. Vendors with shallow offerings might get squeezed out if an AI system can replicate their core features. But for those with deep industry expertise and robust compliance, AI is more of a partner than a threat.
A Shared Path Forward
Ultimately, the question of AI “replacing” SaaS might be the wrong frame. More often than not, we see organizations blend new technologies with existing ones to create better outcomes. AI will change how software is built and used, but it won’t erase the underlying systems of record or domain-specific applications that businesses depend on. Instead, it can make these platforms smarter and more valuable—turning the future into a story of collaboration rather than replacement.
We’ve watched similar transformations before. Each wave of innovation pushes companies to modernize, yet they rarely discard everything that came before. AI may accelerate this process, but it’s unlikely to lead to a wholesale abandonment of SaaS. Instead, expect a dynamic mix of both, reflecting tech’s usual pattern: big changes, but rarely a complete reset.