Let’s ground this conversation and cut through the misunderstood dialect about the impact of artificial intelligence (AI) on the software industry. AI is software and has been part of this category for more than several decades, and yes it has evolved into a new era of capability. Despite the noise from financial markets and headline pundits, the software industry is not being replaced by AI; it is being expanded by it. Software and learning-based systems with models have existed for years, and enterprises have been engaged long before the term generative AI (GenAI) became fashionable. The real discussion should be about how this new generation of AI is evolving enterprise software and what measurable value organizations are willing to pay for in addition to, or instead of, existing approaches.
AI’s role in the software industry spans both software as a service (SaaS) and infrastructure as a service (IaaS), cutting across every layer of engagement, interaction and operations.
AI has introduced real innovation in how intelligence is applied, from software development to the automation of core business processes. The acceleration has followed a clear path. In 2024, the market fixated on GenAI and large language models (LLMs). In 2025, attention shifted to agentic AI, and now the focus is on AI agents and conversational frameworks that engage with people. Foundational LLMs revealed what was possible and proved they could augment long-standing enterprise software and, in some cases, replace it. The value of LLM-based AI software rises sharply when intelligence is embedded directly into business workflows instead of sitting beside them. That value multiplies again when paired with agentic AI that can take actions and make decisions based on data, not just summarize it. Whether those decisions are autonomous or guided by a human in the loop, the efficiency gains are substantial. They become transformative when embodied intelligence is managed as AI agents operating continuously without waiting for human intervention.
The pace of AI innovation has already been absorbed into the SaaS segment where real business applications and platforms live. Traditional enterprise software providers have moved quickly to fuse application platforms with AI platforms, new models and into modern platform architectures. This shift has improved the competitive baseline, not through experiments but through the potential value that can be achieved by an enterprise. Yet too many voices reduce the conversation to claims that AI is a separate market and will exterminate the software industry. Enterprises operate in the real world, where governance, security, compliance and operational continuity matter more than a clever prompt. Proving AI value is not about comparing features to last year’s approach. It is about fitting AI into enterprise architecture, investment strategy and day-to-day operations without breaking the business and while supporting IT processes.
Protecting business continuity demands innovation that can be adopted incrementally, not through reckless rip-and-replace experiments. Technological change at this scale can damage operations when organizational change management (OCM) is treated as an afterthought, and AI magnifies that risk. Software strategy matters, and the choice of approach and provider makes a material difference to outcomes. The wrong path can harm brand trust, disrupt operations and erase years of process discipline. Anyone claiming technology approaches do not matter is simply wrong, and decades of failed transformations have already delivered that verdict.
AI is software. The fundamentals still apply.
Regards,
Mark Smith