Salesforce’s AI Moat: Is the Selloff Pricing in Disruption or a Deepening Fortress?

The market has delivered a dramatic reset for software stocks. The forward price-to-earnings multiple for the iShares Expanded Tech-Software ETF has fallen to 22, a 35% discount to its five-year average of 34. This plunge, driven by fears that AI will disrupt entrenched business models, has brought valuations to levels not seen in over five years. As one analyst noted, the selloff "has helped bring software valuations to reasonable levels" for the first time in a long cycle. For a value investor, this sets up a classic opportunity: a market-wide panic that may be pricing in doom for an entire sector, regardless of individual company strength.

Yet the Buffett/Munger philosophy demands we look past the headline price and question what we're really buying. The reset is not just about revenue fears; it's also about the quality of earnings. A key hidden expense is stock-based compensation, which can distort true profitability. For years, companies have reported adjusted, non-GAAP earnings that strip out this cost, presenting a rosier picture. But as Warren Buffett famously asked, if options aren't compensation, what are they? This accounting practice has become a point of contention, especially as some giants like Nvidia have recently announced they will no longer exclude stock-based compensation from its adjusted operating expenses. For a value investor, this shift is a reminder to scrutinize the economics beneath the adjusted numbers.

The market's indiscriminate reaction is the real red flag. The selloff has priced in disruption as if it's universal and immediate, conflating structurally exposed software businesses with more durable ones. The MSCI World Software & Services Index tumbled by more than 20% this year, signaling a loss of conviction in the entire industry's economic durability. This is where patience and discipline are tested. Volatility is noise; the task is to separate the signal of genuine competitive moat erosion from the fear-driven panic selling.

The bottom line is that the reset creates a fertile field for analysis, not a simple bargain. The low forward P/E is a starting point, but it must be weighed against the hidden costs of growth and the uneven impact of AI. The opportunity lies not in buying the sector, but in identifying the specific companies whose intrinsic value remains intact or is even being overlooked in the broader sell-off.

The Buffett Test: Assessing the Width of the Moat

The AI revolution is a classic test of a company's competitive moat. The market is pricing disruption as universal and immediate, but history suggests the impact will be highly uneven. As Warren Buffett and Charlie Munger would emphasize, the key is not the price of the stock, but the width of the economic moat that protects its earnings over the long term. The selloff has indiscriminately punished the entire sector, conflating structurally exposed software with durable businesses.

This creates a fertile ground for value investors to separate the wheat from the chaff.

The evidence points to a clear bifurcation. Companies providing horizontal application layers, like customer relationship management, face the most direct threat from AI agents that could commoditize development and lower switching costs. The bear case is that if AI can autonomously execute complex workflows, the need for traditional software interfaces diminishes, eroding pricing power. Yet this disruption is not yet a reality for mission-critical systems of record, which are deeply embedded in organizations with authoritative data and compliance frameworks. For these vertical providers, the moat may actually widen as AI integration becomes a premium service rather than a threat.

This is where financial transparency becomes paramount. As the market grapples with uncertainty, investors will prize GAAP earnings that reveal the real impact of AI on business economics. The shift by some giants to include stock-based compensation in adjusted metrics is a step toward clearer visibility. It forces a reckoning with the true cost of growth and the durability of profit pools. In this new environment, a company's ability to compound returns on capital, not just its top-line growth, will define its intrinsic value.

Barclays' analysis of standout performers offers a practical lens. Their top picks-Salesforce, Oracle, and DigitalOcean-each represent a different moat. Salesforce's moat is built on network effects and high switching costs within its CRM ecosystem. The company's AI vision, like its Agentforce platform, is designed to deepen that moat rather than erode it. Oracle's moat lies in its large, high-quality remaining performance obligation base, which provides a visible revenue pipeline and validates its pricing power. DigitalOcean, meanwhile, is building a moat in a niche segment of the AI market, targeting inference workloads with higher margins and less capital intensity than the broader model training race.

The bottom line is that the AI reset is a stress test. It will expose companies with narrow or eroding moats while reinforcing those with durable advantages. For a value investor, the task is to look past the headline fears and assess the width of each company's economic fortress. The market's panic may be pricing in doom for the sector, but the opportunity lies in identifying the specific firms whose moats are not just intact, but potentially strengthened by the new technology.

Financial Health and Growth Trajectory

The market's focus on valuation multiples misses a more critical question: which companies have the financial strength and growth trajectory to compound value over the next decade? The evidence reveals a sector in transition, where established leaders face near-term headwinds but are positioned for a reacceleration, while new entrants target high-margin niches with less capital intensity.

For giants like Salesforce, the path forward hinges on navigating a difficult comparison period. The company has seen growth deceleration due to tough comparisons from the 2020-2021 COVID period. Yet, as Barclays notes, the setup for 2026 looks compelling. The firm expects a reacceleration in core revenue growth, driven by easier comparisons and the initial revenue contribution from its AI products. The valuation supports this view, with shares trading at just 14x EV/FCF for calendar year 2027. This is the kind of price that rewards patience, allowing the company to execute its AI roadmap without pressure for immediate quarterly beats.

Oracle presents a similar story of stabilization and visibility. After a volatile 2025, the company is expected to deliver a more predictable 2026. Its key catalyst is the conversion of a large RPO (remaining performance obligation) base to revenue, which will validate both contract quality and management's gross margin projections for AI workloads. This provides a tangible pipeline of future earnings, a hallmark of durable business models. The current valuation, at 22x CY27 PE, appears attractive for a company with this kind of multi-year growth visibility.

Meanwhile, the growth trajectory is also being reshaped by new leadership and strategic pivots. DigitalOcean, for instance, has appointed a former Oracle Cloud executive as its new Chief Product and Technology Officer to lead its platform roadmap. The company is shifting focus to target the AI inference market, a segment projected to be larger long-term than model training. This niche approach offers a potential multi-year growth story with higher gross margins and less capital intensity than competing with hyperscalers. The company has already pulled forward its growth target to 2026, signaling confidence in its new direction.

The broader sector momentum is also a factor. The Internet Software and Services space is seeing robust IT spending on solutions that support hybrid operating environments. This underlying demand for infrastructure, applications, and security software provides a stable foundation for many players. The Zacks analysis highlights several AI-focused stocks in this space, each with strong growth estimates, suggesting the market is beginning to identify specific catalysts within the sector.

The bottom line for a value investor is to look past the sector-wide selloff and assess each company's unique financial engine. It's about finding the firms where the intrinsic value is being overlooked-not because they are broken, but because their path to future earnings is misunderstood. The opportunity lies in those with the balance sheet strength to invest in AI, the visibility of future revenue, and the strategic clarity to navigate the coming decade.

Catalysts and Risks: What to Watch in 2026

The sector-wide bargain is a setup, not a conclusion. For a value investor, the coming year will be defined by a series of catalysts and risks that will test the durability of individual business models. The market's panic has priced in doom, but the real work is to wait for clear signals that separate the companies whose intrinsic value is being overlooked from those whose moats are genuinely threatened.

The most critical watchpoint is the evidence of AI's economic impact. The bear case hinges on AI agents commoditizing software, reducing development costs, and lowering switching costs. This would directly attack the profit pools of horizontal application providers. Investors must monitor for early signs of revenue cannibalization versus new growth. Does AI integration drive higher customer lifetime value and pricing power, as seen with Palantir's AIP boot camps demonstrating adaptability across logistics and manufacturing? Or does it merely accelerate the commoditization of interfaces? The performance of AI-focused picks like Palantir, HubSpot, and MongoDB will provide real-world data on the sector's ability to compound value in this new regime.

Equally important is the quality of earnings. The market's low valuations are built on adjusted metrics that often exclude stock-based compensation. As one analyst notes, this creates a decades-old accounting debate that is now back in the spotlight. For a value investor, GAAP earnings quality is the true test of management discipline and the sustainability of profit. The trend of companies like Nvidia moving to include this cost in adjusted metrics is a positive step toward transparency. Watch for similar shifts across the sector; a company's willingness to show its full cost of growth is a signal of confidence in its economics.

Finally, monitor the broader IT spending environment. The underlying demand for infrastructure and security software remains robust, as evidenced by robust IT spending on solutions that support hybrid operating environments. This provides a stable foundation for many players. However, the market's loss of conviction in software's durability makes it nearly impossible to forecast terminal value. The catalyst for a re-rating will be companies demonstrating that their AI strategies deepen moats, not erode them, while delivering transparent, GAAP-aligned profitability.

The bottom line is one of patience and discipline. The market is pricing in disruption as universal and immediate, but technology shifts are rarely so simple. The opportunity lies in waiting for the clear signals-the specific metrics and events-that validate a company's long-term compounding potential. Until then, the low price is a starting point, not a reason to act.