A recent TechCrunch article described what some are calling the "SaaSpocalypse," with one investor arguing:

"The barriers to entry for creating software are so low now thanks to coding agents, that the build versus buy decision is shifting toward build in so many cases."

That's not an unreasonable claim.

AI coding agents are improving rapidly. The cost of building internal tools is falling. AI-native startups are replicating features that once took SaaS companies years to develop. The traditional per-seat pricing model is under pressure as AI agents replace human users.

There is real disruption happening.

But I don't believe AI represents the most significant structural shift facing SaaS platforms.

Digital and data sovereignty do.

Yes, AI Is Changing the Industry

Let's acknowledge what's true.

AI is:

• Lowering the cost of building software
• Challenging seat-based pricing models
• Accelerating feature replication
• Creating leverage during contract renewals

Those forces matter. And the article rightly points out that public markets are nervous as AI-native startups move quickly while traditional SaaS incumbents look slower and heavier.

But we should also be realistic.

Building a prototype with AI is easy. Replacing durable, compliant, audited enterprise systems is not.

Enterprises still need:

• Governance
• Auditability
• Workflow durability
• Compliance support
• Long-term operational continuity

AI may reshape business models.

Sovereignty reshapes eligibility.

The Real Question Behind "Build vs Buy"

There's another dynamic at play that isn't getting enough attention.

Over the past decade, the market for self-hosted enterprise software was steadily dismantled. Vendors pushed aggressively toward cloud-only models. On-premise offerings were deprecated. "Managed service" became the only path. In many categories, customers no longer had a realistic buy-and-run-it-yourself option.

So when we see the build-versus-buy equation shifting, it's worth asking:

Is build winning because it's better?

Or because buy no longer includes sovereign, self-controlled alternatives?

In many cases, organizations aren't building because AI made it magically cheap.

They're building because it's the only way to retain control.

The Cloud-Only Constraint

Many of today's dominant SaaS platforms are effectively cloud-only.

No self-hosting.
No infrastructure choice.
No alternative deployment model.
No meaningful control over upgrade cadence.

For organizations facing sovereignty requirements (whether regulatory, geopolitical, or risk-driven) that creates a hard constraint.

If the only buying option is a vendor-controlled runtime that doesn't meet data or digital sovereignty requirements, the remaining option is to build internally.

Not because it's superior.

Because it's viable.

The Erosion of Trust

At the same time AI is disrupting pricing models, something else is happening: erosion of trust.

Organizations are asking harder questions:

• Is our data being used to train AI models?
• Is that usage clearly disclosed?
• Who can access our data under foreign legal systems?
• What happens if a government compels disclosure?
• Are we concentrated inside infrastructure that is a prime geopolitical or criminal target?

Large SaaS providers are efficient and scalable. They are also centralized and jurisdiction-bound. And they are high-value targets.

This isn't hypothetical anymore.

Data sovereignty and digital sovereignty are no longer academic discussions. They are becoming procurement filters and regulatory constraints.

Unlike AI feature competition, sovereignty concerns can disqualify a platform entirely.

AI Disruption Is Commercial. Sovereignty Is Structural.

AI disruption affects:

• Revenue models
• Pricing structures
• Competitive dynamics

Sovereignty affects:

• Legal eligibility
• Regulatory compliance
• Market access
• Operational independence

If your architecture cannot meet sovereignty requirements, you don't renegotiate a contract.

You lose the deal.

Or the market.

We're already seeing governments and municipalities begin to formalize digital sovereignty assessments. Data residency is being scrutinized. Vendor lock-in is being measured. Legal exposure is being evaluated.

That's not hype-driven volatility.

That's structural change.

Where Liferay Fits

This is where deployment flexibility becomes strategic.

With Liferay, organizations can choose:

• SaaS, if that trust model works for them
• Customer-controlled cloud deployments
• Government or sovereign cloud environments
• Fully self-hosted operation

Because Liferay is open source, the source remains available. You are not structurally bound to a single provider's infrastructure. You can choose to remain on a specific DXP version, including long-term support releases, rather than being forced into an upgrade cycle that doesn't align with your governance needs.

That doesn't eliminate the need for good architecture.

But it does eliminate architectural captivity.

In a world where:

• AI is reshaping software economics
• Trust in centralized platforms is under scrutiny
• Sovereignty requirements are tightening

The platforms that preserve choice are the ones best positioned for long-term viability.

Final Thoughts

AI may change how software is built and priced.

But sovereignty determines whether software can be used at all.

If I had to bet on what will shape enterprise platform decisions over the next decade, it wouldn't be AI features alone.

It would be control.

And in that context, the future of SaaS isn't about extinction.

It's about whether "buy" once again includes the option to remain sovereign.