AI Agents: The Evolution, Not the End, of the SaaS Business Model

The narrative that AI agents will render Software-as-a-Service (SaaS) obsolete is being met with significant pushback from the industry's most influential leaders. At the AI Impact Summit 2026 in New Delhi, executives from Salesforce, Tata Consultancy Services (TCS), Infosys, and HCL Technologies converged on a singular theme: AI agents are not the 'SaaS killers' some market speculators suggest, but rather a catalyst for a massive expansion in software complexity and volume. This shift marks a transition from 'vibe coding'—the rapid, AI-assisted generation of simple applications—to a more rigorous, enterprise-grade architecture where governance, observability, and auditability are the primary value drivers.

Arundhati Bhattacharya, Chairperson and CEO of Salesforce India, addressed the speculation head-on, noting that the traditional SaaS model is built on much more than just code. She argued that the true value of a SaaS platform lies in its ability to understand complex workflows, recognize specific customer pain points, and ensure that every action taken by an AI agent is observable and auditable. For venture capitalists and startup founders, this suggests that the 'moat' for future software companies will not be the features themselves, but the infrastructure that allows these features to operate safely and reliably within a regulated enterprise environment. The 'vibe coding' era may lower the barrier to entry for building apps, but it does not lower the barrier for building sustainable, enterprise-grade businesses.

“

Salil Parekh, CEO of Infosys, identified a $300 billion services opportunity emerging from the need to orchestrate these new AI capabilities.

The economic implications of this transition are staggering. Salil Parekh, CEO of Infosys, identified a $300 billion services opportunity emerging from the need to orchestrate these new AI capabilities. As enterprises move away from simple chatbots toward autonomous agents, they require sophisticated orchestration platforms that can integrate foundation models with specialized, domain-specific agents. This creates a massive tailwind for IT services firms and startups that can bridge the gap between raw AI power and measurable business value. The 'impossible' is becoming economically viable, but only through the lens of professional services and rigorous implementation.

From a talent and operational perspective, the role of the software engineer is undergoing a fundamental transformation. K. Krithivasan, CEO of TCS, noted that the industry is moving toward an era where engineers will focus less on writing lines of code and more on high-level architecture and validation. This shift is expected to lead to an explosion in the volume of software produced, rather than a shrinking of the sector. The challenge for the next generation of startups will be managing this increased complexity. As the volume of code grows, the need for 'rigorous validation' becomes the bottleneck, creating a new market for AI-driven testing and quality assurance tools.

However, the path to this agent-driven future is not without its hurdles. C Vijayakumar, CEO of HCL Technologies, pointed out a persistent gap between the capabilities of foundational large language models (LLMs) and the requirements of enterprise-grade performance. Current models, while impressive, often lack the precision and reliability needed for mission-critical enterprise use cases. This 'performance gap' represents a significant opportunity for startups focused on fine-tuning, RAG (Retrieval-Augmented Generation), and specialized foundation models that prioritize accuracy over general-purpose conversational ability.

Looking forward, the consensus from the AI Impact Summit suggests that the SaaS model is entering its most productive phase yet. Rather than being replaced, SaaS platforms are becoming the 'operating systems' for AI agents. The winners in this new landscape will be those who can provide the governance frameworks that allow enterprises to deploy agents at scale without sacrificing security or compliance. For the venture capital community, the focus should shift from 'AI-first' applications to 'Agent-ready' infrastructure and services that can capture a share of the projected $300 billion market expansion.