UiPath Trades at a Discount as AI Adoption Gains Traction in Key Clients

The AI trade is in a state of expectation arbitrage. After a period of record infrastructure spending, the market has shifted from betting on potential to demanding proof. Sentiment is now mixed and skeptical, as investors watch hyperscalers pour capital into data centers with returns still on the horizon.

This dynamic played out clearly with Nvidia's recent earnings. The company posted a genuine blowout quarter, with revenue up 73% year over year and data center demand accelerating. Yet shares fell more than 5% the day the results were released. The numbers beat expectations: EPS of $1.62 on revenue of $68.1 billion topped analyst forecasts. Even the forward guidance was strong, with the company projecting Q1 revenue above Wall Street's estimates. This is a classic "sell the news" moment. The market had already priced in a strong print, leaving little room for a positive surprise to drive the stock higher.

The core of the skepticism is the massive adoption gap. While the hype is about AI transforming every business, the reality is that most companies haven't started using it on a day-to-day basis. According to research, only 18% of businesses are currently using AI. That figure is expected to rise, but it underscores the long runway ahead. The market is questioning whether the current spending spree by hyperscalers is justified by near-term demand, or if it's building infrastructure for a future that is still years away.

The bottom line is that the AI trade is evolving. The era of rewarding any company with "AI" in its investor deck is giving way to one that demands evidence of monetization and a credible path to returns. For now, the market is pricing in potential but waiting for the proof of concept to materialize.

The Fundamental Reality Check: Usage, Disruption, and Diversification

The market's expectation gap isn't just about spending; it's about impact. There's a growing disconnect between the anxiety over AI disruption and the actual financial results for software companies. On one side, the fear is palpable. Mentions of AI disruption on S&P 500 earnings calls nearly doubled from the prior quarter. This is the kind of talk that can spook markets, especially for software firms whose models could be replicated. Yet, on the other side, the fundamentals tell a different story. Fundamental earnings for most software stocks have remained stable. Companies are reiterating guidance, and some, like Salesforce, are even signaling confidence with massive share buybacks. The market may be pricing in a broad, existential threat, but the reality for many firms is a manageable evolution, not an imminent collapse.

This bifurcation is key. AI is not a monolithic force; it's reshaping economic moats. For some, like Microsoft, AI is a powerful amplifier. The company's wide moat is seen as AI-resilient, with its integrated cloud and productivity suite providing a defensible position. For others, the threat is real. Research from Morningstar highlights that AI could downgrade the economic moats of Adobe, Salesforce, and ServiceNow. The expectation gap here is clear: the market is pricing in a wave of disruption that hasn't yet materialized in the earnings of these companies, creating a potential setup for a reassessment.

Nvidia's own story underscores this complexity. While the company is the undisputed engine of the AI buildout, its data center revenue is diversifying beyond its largest customer. CFO Colette Kress noted that hyperscaler revenue increased and remained our largest customer category at slightly over 50% of Data Center revenue, but that growth was "led by the rest of our Data Center customers as revenue diversified." This is a critical point. The market had priced in Nvidia as a pure-play on a handful of mega-hyperscalers. The reality is a more balanced, less concentrated customer base, which reduces a key vulnerability and strengthens the fundamental outlook.

The bottom line is that the AI trade is becoming a game of selective conviction. The hype is about universal disruption, but the proof is in the financials, which show stability for many. The winners will be those whose moats are being fortified, not eroded, and whose revenue streams are diversified away from single, high-risk customers. For now, the market is overestimating the breadth of the disruption while underestimating the resilience of certain business models.

Valuation and Catalysts: Where the Market Might Be Wrong

The expectation gap in AI is now a valuation gap. After years of premium pricing for AI exposure, the market is demanding a discount for the long wait ahead. This creates a setup where some names are being overlooked, not because they lack potential, but because their value is being priced for a distant future.

UiPath exemplifies this dynamic. The automation platform is down 26% year-to-date and trades at just 14x forward earnings. That discount is stark when you consider the fundamentals. The company reported solid growth in its latest quarter, with total revenue up 14% year-on-year and a net retention rate of 107%. More importantly, its core business is already embedded in the most valuable customers, with 90% of clients generating over $1 million in ARR already using its AI products. The market is pricing in a story of slow adoption and high competition, but the reality is a durable platform with sticky, high-value customers. This is a classic case of a stock trading below its intrinsic value because the broader AI narrative has shifted to focus on infrastructure and compute, not the software that manages it.

The key catalyst to close this gap is a tangible acceleration in business AI adoption. The market consensus is stuck on the current 18% usage rate among businesses. The expectation reset will come when that number starts climbing meaningfully, signaling the end of the pilot phase and the beginning of widespread deployment. This would validate the entire infrastructure build-out and force a reassessment of the companies that enable it, from chipmakers to platform providers.

Other near-term catalysts could provide the spark. Earnings reports from key infrastructure players like Micron Technology will offer a direct read on demand for memory chips, a critical component in AI systems. More broadly, policy developments may also shift the calculus. The potential expiration of the Section 122 tariff in July could ease a cost headwind for some AI hardware, improving margins and sentiment. These are not the moonshot catalysts of the hype cycle, but they are the concrete events that can trigger a re-rating when they align with the underlying adoption trend.

The bottom line is that the market's skepticism is rational, but it may be overdone. When adoption finally accelerates, the valuation disconnects we see today-like UiPath's discount-will likely snap shut. For now, the opportunity lies in identifying the durable platforms and the specific milestones that will prove the skeptics wrong.