Gutwillig examines whether agentic AI and enterprise software are on a collision course — analyzing how LLM-powered coding agents are reshaping Build vs. Buy dynamics, SaaS valuations, and the future of platforms like Salesforce, SAP, and Workday.
From the original report
Selected passages from the original publication. Historical observations reflect the report's publication date, not current market advice.
If you own the leading enterprise software vendors in your portfolios, it is highly likely that you are among those of us experiencing “a rollercoaster ride” on your holdings over the past 12 months and may be interested in what’s been driving the volatility as well as where the markets may be headed as it relates to enterprise software vendors (also often referred to as Software as a Service or SaaS vendors).
In this month’s CTO Newsletter, we explore how several of the enterprise software leaders are being impacted by AI and how these firms are evolving their individual business strategies (from both product portfolios and go -to-market approaches) to address growing client base interest in intelligent automation of their core business processes, often in a highly “tailored” manner specific to the unique characteristics of the industry and business processes.
From our perspective at Gideon, the enterprise software sector appears to be in a correction phase after excessive AI-driven optimism in mid-2025, with valuations resetting to more sustainable levels as the market demands clearer evidence of profitable AI revenue generation.
Today, the markets appear to be at an inflection point in enterprise software. Large Language Models (LLMs) and the rapid emergence of agentic AI are no longer experimental overlays or “bolt-on” capabilities integrated with APIs, they are now beginning to rewire how value is created, delivered, and monetized across the enterprise software stack. While both traditional workflow automation and agentic AI focus strongly on transactional processing efficiency, AI agents are also capable of intelligence, adaptability, and continuous learning. They can take autonomous, goal-directed actions and process and optimize workflows at an unprecedented rate, without the latency issues associated with more traditional methods. These AI-powered, interconnected agents embedded into workflows will even be able to adapt dynamically to changes in the environment, detecting and fixing issues independently and then moving to prevent them from happening again.
For educational purposes only; not individualized investment, tax, or legal advice. Investing involves risk, including loss of principal.
The original PDF is available as an optional companion to this public article. Use the download form below to access the document.