SaaS Evolution Accelerates With AI As Cloud Investment Surges Despite Industry Sell Off

Stephen Howe

The debate intensified in early 2026 when SaaS company valuations saw a massive sell-off. Nearly $1tn in market value was erased from software stocks, as investors reassessed growth expectations amid rapid AI adoption and structural change in enterprise software (Forrester).

Yet despite volatility in public markets, broader industry forecasts indicate that worldwide spending on public cloud services is forecast to double in size by 2028 from its $805bn in 2024 (IDC Worldwide Software and Public Cloud Services Spending Guide).

In other words, SaaS is not shrinking. It is scaling.

For Stephen Howe, director at Times 3 Technologies (T3T), the narrative of ‘SaaS extinction’ misses a critical point about how enterprise systems actually function.

“Is SaaS dead with AI? Absolutely not,” says Howe. “For AI to deliver real value, you need cloud-connected ecosystems, trading partners, and shared data environments. If you are still sitting in a closed on-premise system with no external connectivity, you simply cannot unlock those benefits.”

He argues that AI does not replace enterprise software systems like Enterprise Resource Planning (ERP), Human Capital Management (HCM), Customer Relationship Management (CRM), or IT Service Management (ITSM), it depends on them.

“These systems are not optional tools. They are systems of record,” he explains. “AI needs structured, trusted data to be meaningful. You cannot train intelligence in isolation and expect it to outperform years of accumulated enterprise logic.”

Increasingly, AI is being embedded directly into SaaS platforms. Industry research shows that 60% or more of enterprise SaaS products now include embedded AI capabilities, with AI fast becoming a required layer in SaaS platforms from predictive analytics to generative copilots and automated workflow engines (Founders Forum Group).

This shift is turning SaaS platforms into intelligent systems rather than static applications.

However, Howe stresses that AI’s real value lies not in isolated automation, but in shared intelligence across organisations.

“There are two layers of AI emerging,” he says. “One is internal process optimisation and how a business runs its own workflows. The other is AI-driven automation across businesses, where learning and insights are shared. That’s where vendors should be enabling collaboration and best-practice frameworks.”

He points to practical examples already emerging in HR and payroll systems, where AI copilots can respond to employee queries such as leave balances or payslip explanations in real time, reducing administrative load and improving employee experience.

“What used to be a data-capture role is becoming an insight-driven role,” Howe adds. “Instead of searching for information, employees are being prompted with relevant insights while they work like identifying a customer who hasn’t paid or suggesting a relevant discount based on prior purchasing behaviour.”

This shift, he argues, fundamentally changes how organisations operate. AI is not replacing ERP or payroll systems, it is removing friction between users and data.

At the same time, South Africa remains in what Howe describes as an “early adoption phase.” While larger enterprises are actively exploring AI integration, many organisations are still experimenting with how best to incorporate it into their operations.

A key strategic question, he says, is whether businesses should attempt to build proprietary AI systems or instead leverage global software ecosystems already embedding AI at scale.

“The smarter approach is to build on established enterprise platforms that already encode best practice,” Howe says. “That ensures governance, compliance, and scalability are built in from the start.”

This aligns with broader industry thinking: as AI becomes more pervasive, data sovereignty, governance, and auditability are increasingly viewed as strategic capabilities and competitive advantages particularly in regulated industries, where trusted data and compliance-ready architectures are essential for successful AI deployment.

Ultimately, while SaaS may be undergoing one of its most significant transformations since its inception, its foundation remains intact.

AI is not dismantling enterprise software. It is rewiring it. SaaS is not dead. It is becoming something far more powerful.