How the rise of AI-native software could give SMBs enterprise-level power

imaginima/iStock/Getty Images Plus

Follow ZDNET: Add us as a preferred source on Google.

ZDNET's key takeaways

• AI-native software providers will challenge traditional SaaS. 
• Rising AI infrastructure costs present a challenge.
• More native-AI vendors mean more choices for smaller businesses. 

Prepare to see more solutions from 'AI-native' software companies in the coming months and years. This trend means new interfaces, different cost considerations, and new ways of building and working with applications. Just as cloud-native vendors flooded the market a decade ago with Software-as-a-Service (SaaS) and Platform-as-a-Service (PaaS) offerings, we may see a glut of new AI-based applications. 

That's the word from a new study by consultant Deloitte. It's not that SaaS vendors are going away anytime soon. Right now, the top 10 SaaS providers account for more than half of the software market's capitalization. The overall market grew by 11% between 2024 and 2025, up from $3.6 trillion to $4 trillion. 

Also: 90% of AI projects fail - here are 3 ways to ensure yours doesn't

But market dynamics are rapidly changing. SaaS and traditional software companies are already under pressure to replace per-license pricing with results-driven pricing. 

"Competition will heat up as AI-native challengers begin to chip away at market leaders across business processes and create new market segments that were previously unaddressed by software," the study's authors predicted. 

"In addition, new entrants are rapidly growing and disrupting the market with leaner operating models." 

Also: 5 ways you can stop testing AI and start scaling it responsibly in 2026

This shift has interesting implications for developers, engineers, designers, and product managers, who will need to reorganize or transform their operations and skillsets.  

So, what service areas are changing? The study points to several areas, including the following:

• The growth of AI as an interface: "Over the next few years, AI-powered systems are expected to increasingly act as the primary interface across multiple software applications," the report stated. "Fierce competition among software companies to be the primary interface layer is expected to keep customers within their platforms and give providers access to agent telemetry."
• The rising prominence of AI orchestration platforms: Traditional applications are evolving into collections of autonomous AI agents. AI orchestration platforms will be required to monitor and manage agents.
• The struggle to keep a lid on compute costs: The costs associated with AI infrastructure will likely "squeeze margins for software companies in 2026," the study predicted. "The additional costs from using large language models, investments in new agentic products, and hybrid pricing could pressure future revenues and margins."

This new environment will be especially advantageous to small and medium-sized businesses (SMBs). "AI-first software lets SMBs operate like enterprises, delivering advanced capabilities at a fraction of the cost," Ayo Odusote, software and platforms leader at Deloitte, told ZDNET. 

"With increased competition from AI natives, buyers, including SMBs, will have more options and therefore more power, potentially leading to lower costs. Plus, low-code and no-code platforms mean that you don't necessarily need access to expensive technical talent to create software."

Still, smaller enterprises "will need to manage upfront investments carefully, particularly around data readiness, integration, and governance," Odusote cautioned. "The long-term savings and productivity gains are real, but the cost advantage comes from disciplined deployment and measurable business outcomes, not experimentation alone."

Also: AI isn't getting smarter, it's getting more power hungry - and expensive

The rise of a new class of AI-native vendors also means an embarrassment of riches -- with many new choices. "Because there is more competition in the space and more options, it might also make it harder for SMBs to decide on software vendors," said Odusote. "Do they go with an established vendor, or look to someone new who could offer more innovation at potentially lower cost and take a risk?"

There is intensifying competition between incumbents and AI-native challengers, "many of whom are focused on highly specialized use cases," he stated. "That dynamic is creating more choice, particularly for SMBs that may prefer modular, targeted solutions rather than large, monolithic platforms."

We're seeing the formation of an "AI-driven software layer across industries," Odusote said. "Smaller startups are capitalizing on niche workflows and vertical-specific AI applications, which could result in a more diverse and innovation-rich marketplace. For SMBs, that can translate into more tailored solutions and competitive pricing."

Also: 5 ways rules and regulations can help guide your AI innovation

The adoption of AI solutions also requires workforce evolution, he continued. "The transition to AI-first is as much about workforce evolution as it is about technology. Organizations scaling AI successfully are redesigning roles and building AI literacy across the enterprise." 

Key skill areas required include data management, vendor evaluation, workflow redesign, and cross-functional collaboration between business and technical teams.