House View — UiPath (PATH): The AI-chain Play the Market Gave Up On
UiPath just turned GAAP-profitable for the first time while its stock trades near multi-year lows. The market is paying for the layer that builds intelligence and almost nothing for the layer that governs it. Our fundamental AI-chain position.
Ballad Markets · Desk View · July 2026
There is a specific kind of setup this desk waits for: a company whose fundamentals are inflecting upward while its share price is making multi-year lows. Not a broken business trading cheap — a repairing business trading cheap. UiPath is that setup today.
The stock changes hands near $10, down from roughly $12.15 earlier this month and within touching distance of its 52-week low of $9.20. Its all-time high, set in May 2021, was $85.12. The market has spent three years pricing UiPath as a melting ice cube.
The financials no longer support that story.
What actually happened last quarter
UiPath reported Q1 fiscal 2027 (quarter ended April 30, 2026) on May 28:
| Metric | Result |
|---|---|
| Revenue | $418M, +17% YoY |
| ARR | $1.901B, +12% YoY |
| Net new ARR | $49M |
| Net retention | 109% |
| GAAP operating income | +$28M — first positive quarter in company history |
| Non-GAAP operating income | $92M |
| Non-GAAP gross margin | 83% |
| Adjusted free cash flow | $130M |
| Cash & securities | $1.42B, no debt |
| Buyback | $244M of Class A stock repurchased in the quarter |
The line that matters is the fifth one. UiPath turned GAAP-profitable for the first time. For a company the market had written off as a perpetual cash-burner, that is a change in kind, not degree — and it happened while revenue still grew 17%.
Management’s own guidance carries it forward:
- Q2 FY27: revenue $395–400M, ARR $1.929–1.934B, non-GAAP operating income ~$75M
- Full-year FY27: revenue $1.776–1.781B, ARR $2.058–2.063B, non-GAAP operating income ~$430M
That full-year operating income figure against ~$1.78B of revenue implies roughly a 24% non-GAAP operating margin — from a company that ran near breakeven two years ago. Meanwhile it holds $1.42B in net cash against a market capitalisation of roughly $5.5B, meaning a quarter of the market cap is cash.
Why the price is where it is
We are not going to pretend the bear case doesn’t exist. On July 23, UBS cut its price target to $12 from $13 and kept a Neutral rating, reflecting broader scepticism about automation demand. That scepticism is the reason the entry exists, and it rests on a real question: does classic RPA get commoditised by general-purpose AI?
The honest answer is that it is a genuine risk. Microsoft bundles Power Automate. Foundation-model vendors keep absorbing adjacent workflows. If enterprise automation collapses into whatever the model providers ship for free, UiPath’s moat erodes.
Our read is that this conflates two different things. A language model can draft a process. It cannot, on its own, govern one — permissions, audit trails, deterministic retries, compliance evidence in a regulated bank. Orchestration is where the durable value sits, and orchestration is what UiPath sells.
The AI-chain thesis
The reason PATH enters our watchlist as a fundamental AI-chain position rather than a bounce trade: it is levered to enterprise AI deployment, not AI capability.
The market has paid up enormously for the layer that builds intelligence — chips, compute, foundation models. It has paid almost nothing for the layer that puts that intelligence to work inside a regulated enterprise and proves it did what it claimed. Every agent an enterprise deploys needs to be scheduled, permissioned, monitored and audited. That is not a model problem. That is an orchestration problem.
Two data points suggest the shift is already monetising. On UiPath’s Q1 call, management noted AI-attached deals sold materially larger than those without. And the customer evidence is concrete rather than theoretical:
- Suncoast reviewed 155,000+ checks with agentic automation, stopped 400 fraud attempts, and prevented $3.3M in losses
- USI processes data 80% faster
- Federal Bank merged 250 records an hour, continuously — work that previously consumed a full employee day for 200–300 records
The named enterprise base spans regulated and industrial sectors: Generali Investments, Hiscox, HUB, WEX, Libra Internet Bank, Omega Healthcare, DTE Energy, Continental Resources, Tetra Pak, Woolworths, SOCAR, TAURON, One NZ.
These are unglamorous companies. That is the point. This is where automation gets bought with real budgets and renewed on real ROI.
The market
Third-party estimates put RPA and hyperautomation at roughly $22.6B in 2025 scaling toward $110B by 2034 — a high-teens compound growth rate. Gartner has placed UiPath in the Leaders quadrant for six consecutive years.
We would treat the market-size figure as directional rather than precise; forecasts nine years out are scenarios, not facts. The relevant point is simpler: the category is growing, and the category leader trades at roughly 3× forward sales while generating cash.
The desk’s framework
Horizon: 3–5 years. This is a position, not a trade. The thesis needs multiple quarters of margin expansion to be repriced, and that is measured in earnings reports, not weeks.
| Scenario | Revenue CAGR | Op. margin | Framework value |
|---|---|---|---|
| Bull | ~25% | ~27% | $25–28 |
| Base | ~15% | ~24% | $14–16 |
| Bear | ≤10% | ~20% | $8–9 |
The asymmetry is what earns the position. From roughly $10, the base case is a meaningful gain, the bull case is a multiple of the current price, and the bear case is bounded — partly because $1.42B of net cash puts a floor under a business that no longer burns money.
What invalidates this: two consecutive quarters of decelerating net-new ARR, a reversal of the GAAP profitability, or net retention breaking below ~105%. Any of those means the operating leverage story broke, and the position should be reconsidered on the evidence rather than defended.
What we are watching next
Q2 FY2027 results are scheduled for September 8, 2026, after the close. That is the next real test. The bar management set is revenue of $395–400M and ARR of $1.929–1.934B.
We hold a constructive view into that print — the last four quarters have shown consistent operating discipline, and guidance has been met. But we will state plainly what we do not know: no one can predict an earnings result, and anyone who tells you they are certain of one is selling something. The position is sized for a thesis that plays out over years, precisely so that a single quarter does not decide it.
PATH is added to the Ballad house view as our fundamental AI-chain watchlist position. CHESKO’s desk agents will run their debate on it and publish their independent read — including, if they find it, the case against.
Ballad Markets publishes its house view for research and educational purposes. This is market commentary, not personalised investment advice, and it is not a recommendation to buy or sell any security. We may hold positions in the instruments discussed. Markets carry risk of loss; do your own work and consider your own circumstances.