Bill McDermott is not a man who does things quietly.
So when he stood at ServiceNow’s annual Knowledge 2026 conference in Las Vegas earlier this month and told thousands of customers, partners, and journalists that he regularly tells Nvidia’s Jensen Huang “we’re going to be a trillion-dollar company” — with Huang sitting right there on stage beside him — it landed exactly the way McDermott intended. Half the room laughed. The other half started doing math.
The math is genuinely humbling. ServiceNow’s market cap sits at roughly $98 billion today. A trillion dollars is ten times that. McDermott is giving himself until 2030 to get there. And in a move that separates genuine conviction from executive cheerleading, he has tied his own compensation to hitting that target. If ServiceNow doesn’t reach $1 trillion in market cap by 2030, he personally pays for it.
That is either one of the most confident bets in the history of enterprise software, or a very expensive lesson in the limits of ambition. Let’s look at which one it actually is.
The Story He Told Jensen and What It Reveals
McDermott shared the anecdote during a media roundtable at K26 on May 5, and it is worth unpacking beyond the punchline.
“I always tell Jensen about the trillion-dollar company,” he said. “When I first saw Jensen at the Knowledge event, Nvidia’s market value was $550 billion. After the event, they announced earnings and it became $1.3 trillion. Jensen came back to Knowledge two years later, and following the next earnings release, it topped $2 trillion. We went to the Dell event, where I did a keynote with Michael. After that event, it reached $3.2 trillion. It’s now unimaginable. Today it’s $5 trillion. I told him that when we meet on the next Las Vegas stage, it will be $6 trillion. Jensen replied, asking if I wanted a trillion-dollar good-luck charm.”
The room laughed. But what McDermott was really doing — and he knows exactly what he’s doing — was positioning ServiceNow inside the same narrative arc that made Nvidia one of the most valuable companies in human history. Nvidia didn’t plan to be worth $5 trillion. It built the infrastructure that turned out to be indispensable to the most important technology wave in a generation. McDermott’s argument, stripped of the stagecraft, is that ServiceNow is doing the same thing: building the platform that enterprises cannot function without once AI agents become the dominant way work gets done.
That is a serious argument. It deserves serious scrutiny.
The Business Underneath the Ambition
First, the numbers that actually matter — because McDermott is not making this argument from a standing start.
When he joined ServiceNow as CEO in 2019, the company was doing $3.46 billion in annual revenue. By 2025, that number had grown to $13.3 billion. The company is tracking to surpass $15 billion this year. At K26, management unveiled a forecast committing to at least $30 billion in revenue by 2030 — a doubling of current revenue in four years. That would make ServiceNow one of the largest enterprise software businesses on earth.
The stock, however, tells a different story right now. Shares are down 39% year-to-date and more than 55% from their 52-week high. The Wall Street narrative punishing ServiceNow isn’t about the revenue numbers — those are strong. It’s about a fear McDermott has a name for: the Saaspocalypse.
The Saaspocalypse theory goes like this: generative AI will make traditional SaaS subscriptions unnecessary. Companies will build internal AI tools that replace the workflows they were previously paying ServiceNow, Salesforce, and others to manage. The SaaS model, in this reading, is being disrupted by the very technology wave those companies are trying to ride.
McDermott’s response is pointed. “AI is a tailwind, not a headwind, to ServiceNow’s business,” he said in an interview with Fortune. “This is Saaspocalypse nonsense.” His counter-argument is that as enterprises deploy more AI agents, they need more infrastructure to manage them — not less. ServiceNow, in his telling, is not a product that AI replaces. It is the platform that makes AI work at enterprise scale. The more agents a company deploys, the more it needs ServiceNow.
K26: What ServiceNow Actually Launched
Setting aside the trillion-dollar theater, K26 was a substantive product conference that laid out exactly how ServiceNow intends to win the AI agent management market.
The centerpiece was Action Fabric — a new subscription model and integration layer designed to connect with any AI model, regardless of vendor. “We are now completely open,” McDermott said in his keynote. “Whatever AI you choose to use, we will help you make the best choice and run it better.” This is a significant strategic shift. Rather than competing against foundation model providers like OpenAI, Anthropic, or Google, ServiceNow is positioning itself as the neutral layer above them — the governance and execution platform that enterprises use to actually deploy and manage those models at scale.
ServiceNow’s Big Bet on AI Agent Infrastructure
Alongside Action Fabric came the full launch of AI Control Tower, which gives companies real-time visibility into every AI agent, workflow, and model running across their enterprise. Think of it as a command center for AI operations. As organizations start running dozens — and eventually hundreds — of AI agents across departments, the chaos of managing them becomes a genuine operational problem. ServiceNow is betting enterprises will pay for the platform that brings order to that chaos.
Then there was Project Arc, developed jointly with Nvidia. It gives enterprises a secure sandbox for deploying autonomous agents — what Huang called an “Open Shell” during the keynote. Every AI agent gets identity management, access controls, and governance policies, governed the same way human employees are. As Huang put it on stage: “You have human agents and AI agents, and they basically govern the same way.” That framing matters because it positions AI agent management not as a technical problem but as a workforce management problem — one ServiceNow has been solving for enterprises for years.
ServiceNow Otto rounds out the K26 launches: conversational AI, autonomous workflows, and enterprise search integrated into a single product.
The market context for all of this: according to IDC, the global number of AI agents is expected to grow from 28.6 million in 2025 to 2.2 billion by 2030. If that forecast is anywhere near accurate, the management infrastructure required to govern all of those agents is a market ServiceNow is almost uniquely positioned to serve.
Jensen Huang’s Presence and What It Actually Means
Huang’s appearance at K26 was not symbolic. This was his third consecutive year attending the conference. Nvidia runs its own employee workflows on ServiceNow — the partnership is operational, not just a stage appearance. When Huang shows up and says AI creates jobs rather than destroying them, it carries weight that a CEO of a smaller company simply cannot manufacture.
His message to corporate leaders at K26 was characteristically blunt: “The time has come to use AI to do the work.” He pushed back against the narrative that AI destroys employment, framing the opposite: as long as companies have more ambition than they can currently execute on, AI creates jobs by expanding what’s achievable. He described opportunity growing across all five layers of what he calls the AI economy: energy, chips, infrastructure, models, and applications.
For McDermott, having Huang on stage is a strategic asset. Nvidia’s trajectory from $550 billion to $5 trillion is the proof of concept for his trillion-dollar thesis — not that ServiceNow will follow the identical path, but that infrastructure companies at the center of major technology waves can achieve valuations that look absurd until they don’t.
The Real Challenge McDermott Faces
None of this means the trillion-dollar target is likely. It means it is not absurd.
Getting from $98 billion to $1 trillion requires market cap growth that very few companies in history have achieved in a four-year window. For Action Fabric to succeed, the subscription model must generate meaningful new revenue. The broader AI agent management market also needs to expand at the pace IDC projects. Just as importantly, the “Saaspocalypse” narrative — which has already erased more than half the stock’s value from its highs — must ultimately prove wrong.
McDermott bought $3 million of ServiceNow shares himself in February, at a point when the stock was under serious pressure. More than 90% of ServiceNow employees are also buying company stock, he said. He has walked over “tougher challenges than that on my way to a fight,” in his own words. The personal financial exposure, at least, is real.
Conclusion
ServiceNow in May 2026 is a company with genuine revenue momentum, a coherent AI strategy, a legendary technology partner on stage beside its CEO, and a stock that has been punished severely by a market narrative it cannot yet disprove. Whether Action Fabric, AI Control Tower, and Project Arc become the enterprise backbone of the AI agent era — or whether the Saaspocalypse turns out to have more teeth than McDermott thinks — will determine whether the trillion-dollar target is remembered as audacious vision or expensive overconfidence.
The bet has been placed. The compensation is on the line. The good-luck charm from Jensen Huang has been accepted. Now the company has to build.
References
- Seoul Economic Daily — “ServiceNow CEO Aims to Join Trillion-Dollar Club Like Nvidia” (May 8, 2026) — https://en.sedaily.com/international/2026/05/08/servicenow-ceo-aims-to-join-trillion-dollar-club-like-nvidia
- Fortune — “Why CEO Bill McDermott says ServiceNow’s 39% stock crash is Saaspocalypse ‘nonsense'” (May 8, 2026) — https://fortune.com/2026/05/08/servicenow-ceo-bill-mcdermott/
- Fortune — “Jensen Huang on why ‘agentic’ will rewire a $50 trillion economy” (May 6, 2026) — https://fortune.com/2026/05/06/jensen-huang-servicenow-bill-mcdermott-agentic-ai-robos/
- Seoul Economic Daily — “AI Is the Answer to 50 Million Worker Shortage” (May 7, 2026) — https://en.sedaily.com/international/2026/05/07/ai-is-the-answer-to-50-million-worker-shortage-servicenow
- Business Chief — “NVIDIA CEO: Have More Ambition Than Your Company Can Handle” — https://businesschief.com/news/nvidia-ceo-have-more-ambition-than-your-company-can-handle
- BigGo Finance — “ServiceNow CEO Tells Nvidia’s Jensen Huang ‘We’re Going to $1 Trillion Too'” — https://finance.biggo.com/news/aAFfBJ4BYH_ypPqORDjQ
- WDC TV News — “ServiceNow CEO: Saaspocalypse Nonsense and Trillion Dollar Ambition” — https://wdctv.news/servicenow-ceo-bill-mcdermott-interview-saaspocalypse-nonsense-and-trillion-dollar-ambition/
- IDC — Global AI Agent Forecast 2025–2030 (cited in K26 press coverage)
FAQs
Q1: What is ServiceNow and what does it actually do?
ServiceNow is a cloud platform that large companies use to manage internal operations — IT support, HR processes, customer workflows, and increasingly, AI agent governance. It is used by thousands of global enterprises. Annual revenue reached $13.3 billion in 2025, making it one of the largest enterprise software businesses in the world.
Q2: Why has ServiceNow’s stock dropped so much despite strong revenue growth?
The market is pricing in a fear that generative AI will make traditional SaaS platforms obsolete — that enterprises will build internal AI tools instead of paying for subscriptions. This narrative, which McDermott calls the “Saaspocalypse,” has weighed heavily on the entire enterprise software sector. ServiceNow shares are down 39% year-to-date and over 55% from their 52-week high as of May 2026.
Q3: What is “Action Fabric” and why does it matter?
Action Fabric is a new ServiceNow subscription model and integration framework that connects with any AI model or platform, regardless of vendor. It positions ServiceNow as a vendor-neutral orchestration and governance layer for enterprise AI — meaning companies use it to manage AI regardless of which AI tools they’ve chosen to deploy.
Q4: What is the relationship between ServiceNow and Nvidia?
The relationship is both commercial and strategic. Nvidia runs its internal employee workflows on ServiceNow. Project Arc, announced at K26, is a joint product that provides enterprises with a governance framework for deploying autonomous AI agents. Jensen Huang has appeared at ServiceNow’s Knowledge conference for three consecutive years.
