Can Software Survive AI? Why the Sector is Entering a New 'Digestion' Phase

The narrative surrounding the software industry has shifted dramatically over the last three months. What began as a period of unbridled enthusiasm for generative AI has morphed into a climate of caution, with investors increasingly concerned about "AI disruption risk."

But is the software sector facing an existential threat, or simply a necessary evolution? According to Gabriela Borges, Software Sector Analyst at Goldman Sachs Research, the answer lies in how incumbents leverage their unique advantages to navigate a cooling growth environment.

The End of the "Growth at Any Cost" Era

For investors who cut their teeth during the 2020–2021 boom, the current landscape feels jarring. Sector growth has decelerated from over 20% to approximately 10% today. This has ushered in a "digestion" period for enterprise tech, where the focus has moved from headcount-heavy expansion to margin discipline.

Borges highlights that companies like Okta have become poster children for this shift, successfully increasing free cash flow margins by roughly 20% over an 18-month period. This pivot is critical because it is changing the investor base. By moving away from non-GAAP metrics that often mask the dilution of stock-based compensation, software companies are finally attracting "Growth at a Reasonable Price" (GARP) and value-oriented investors who previously avoided the sector.

The Moat: Domain Experience vs. AI Wrappers

The primary fear among investors is that AI-native startups—or tools like Anthropic’s Claude Code—will render legacy software obsolete. However, Borges argues that incumbents possess a distinct, defensible advantage: domain experience and proprietary data.

"CrowdStrike has been collecting data on endpoints—mobile phones, laptops, servers—for 10-plus years," Borges notes. "They have a human in the loop doing reinforcement learning and infusing their own tradescraft into that data."

For incumbents, the challenge is to prove that their AI agents—such as Microsoft’s Copilot or Salesforce’s AgentForce—can deliver superior outcomes compared to third-party AI wrappers. If they can demonstrate that their deep context and data lead to better results, their moats will remain intact.

The Three-Step Playbook for Survival

To separate the winners from the losers, Borges suggests a three-step checklist for evaluating software leaders:

1. Modernize the Stack: Companies must address legacy technical debt. If the underlying architecture is 15–20 years old, it will struggle to support modern, fluid AI data flows.
2. Strategic M&A: Don't rely solely on organic R&D. Leaders should use their balance sheets to acquire private AI startups that fill roadmap gaps. Salesforce, for example, has completed over 10 acquisitions in the last 12 months to bolster its capabilities.
3. Monetization Discipline: Success is no longer just about "AI adoption." It is about the ability to track usage and price for it. Investors should monitor KPIs like bookings, billings, and the percentage of seats monetizing AI functionality as leading indicators.

"Drinking Their Own Champagne"

Perhaps the most optimistic development is the potential for software companies to decouple their operating expenses from headcount. By using AI to improve their own internal productivity—"drinking their own champagne"—software firms can shift their cost structure.

"OPEX growth going forward may actually be more tied to the cost of compute, which we know is coming down, rather than the cost of hiring and retaining talent," Borges explains. This shift could provide the valuation support necessary for the sector to stabilize and eventually re-rate.

Key Takeaways for Investors

• Look for Fundamental Stability: The "wait and see" mode is ending. Pay close attention to earnings reports where companies with strong data points (bookings/billings) are being rewarded by the market.
• Prioritize GAAP Earnings: Avoid companies that rely heavily on non-GAAP accounting to hide the dilution of stock-based compensation. The market is increasingly demanding high-quality earnings.
• Monitor the Private Ecosystem: The best public company investors are now tracking the private startup landscape. If a company isn't innovating, check if they have the M&A firepower to buy the innovation they lack.
• Watch the "Fast Followers": You don't need to be the first to market. Companies like Workday, Intuit, and Palo Alto Networks are well-positioned to leverage their scale to integrate AI faster and more reliably than smaller, unproven startups.

While the competitive intensity is undeniably higher, the software sector is not disappearing—it is maturing. For the disciplined investor, this transition offers a rare opportunity to buy into high-quality platforms that are finally prioritizing profitability alongside innovation.

---

Disclaimer: This article is for informational purposes only and does not constitute financial advice. Always conduct your own due diligence before making investment decisions.