Software Companies Aren't Just Selling Code

The prevailing bear case for software companies suggests that as AI commoditizes the act of writing code, the proprietary value of software evaporates. If an LLM can generate a bespoke application in seconds, the logic goes, the "moat" of the SaaS business model is breached.

This view ignores the fundamental reality of enterprise procurement: companies don't just buy code, they buy a transfer of risk. A business doesn't merely pay for a functional UI; it pays for a legal counterparty. In enterprise operations, there is a legal infrastructure, from compliance to security protocols to indemnification, that an AI agent simply cannot inhabit. You cannot sue an in-house DIY, built-over-the-weekend application and you certainly cannot extract liquidated damages from a local LLM when a system failure halts a production line.

Ultimately, and for years to come, the buying and selling of services requires a human and adheres to global legal norms. While AI can optimize the "factory" of software production, it cannot replace the trust blanket provided by a contractually bound vendor.

Software is a commitment to a never-ending journey of maintenance and security. It is also a professional partnership backed by insurance and accountability. For enterprises, the peace of mind afforded by a sue-able counterparty is worth more than the cost of the code itself. And will be for some time.