ServiceNow Bundles AI Into Every Product and Bets That Context Is the Missing Piece
We keep hearing that enterprise AI spending is up in many organizations, and the results are not remarkable. On April 9, 2026, ServiceNow announced it will no longer sell AI as an add-on, instead restructuring its commercial model to focus on AI and launching new features that let AI agents make organizational decisions without frequent human oversight.
In the previous model, ServiceNow customers paid for a platform and then paid again to activate AI on top of it. ServiceNow announced changes to their packaging. All products now (in April 2026) ship with AI, Workflow Data Fabric for cross-system data access, AI Control Tower for agent governance, Moveworks as the employee-facing conversational layer, and Process Mining included by default.
The new three-tier structure runs from Foundation through Advanced to Prime. The Foundation tier covers AI assistance for human workers. The Advanced tier automates complete workflows end-to-end. The Prime tier deploys autonomous AI specialists that execute role-based work independently. Pricing uses a hybrid model; subscription commitments paired with usage tokens, called “assists,” where token consumption scales with the complexity of the AI action rather than by seat count.
ServiceNow named the competing approaches directly. Pure per-user pricing front-loads cost before value is proven. Pure consumption pricing makes multi-year planning difficult for finance leaders. Buyers should still build their own cost models before committing to the Prime tier, where autonomous AI activity could drive token consumption that is hard to forecast until you have real usage data.
The Midmarket Opening Is the Right Call
ServiceNow introduced Enterprise Service Management Foundation, a bundled offer for midsize organizations covering IT, HR, legal, finance, procurement, and workplace services. The target is companies that run multiple disconnected point tools and want enterprise-grade service management without an enterprise-scale implementation project. ServiceNow claims a live deployment in approximately 30 days using AI-guided setup. This would be a radical departure from the typical timeframes we see at Info-Tech for ServiceNow implementations and should be taken with a grain of salt.
Midmarket organizations have historically sat outside ServiceNow’s addressable customer base, priced out or overwhelmed by implementation complexity. A faster, lower-complexity entry point on a shared architecture is the right move. A customer can scale into higher-value automation tiers without migrating to a different platform, which is exactly the kind of expansion motion that changes a vendor relationship from transactional to strategic.
Context Engine Solves the Problem Agents Keep Hitting
Amit Zavery, president and chief product officer, made a point during the briefing that cuts to the heart of why most enterprise AI pilots stall. The problem is not the model. The problem is that the model has no idea how your organization actually works. It does not know that this employee’s role qualifies for an exception, that this vendor has a history that should trigger a review, or that every similar request for the past six months was approved by precedent. Without that context, every edge case needs a human.
Context Engine is ServiceNow's answer to that problem. It connects five data layers – identity and access relationships, asset dependencies, organizational knowledge and policy, decision traces, and action history – into a unified layer that AI agents query at decision time. Every resolved ticket and approved request feeds back into the graph. The system gets smarter about how a specific organization operates with every action it takes. ServiceNow becomes a better retrieval system. It has organizational memory that builds over time and makes autonomous decisions progressively safer and more precise.
Context Engine is in preview with select customers. Full availability details have not been announced.
“Build Anywhere, Govern on ServiceNow” Is Their Announcement for Developers
Starting April 15, developers can build applications and agents using any major AI development environment, including Claude Code, Cursor, OpenAI Codex, and Google Antigravity, and deploy directly to the ServiceNow platform. The ServiceNow software development kit and Build Agent skills package platform knowledge into those environments so they’re production-ready from the first line of code.
Usually applications that get built fast and deployed outside IT visibility are red flags, as the governance is missing. ServiceNow is positioning itself as the destination that captures that build-anywhere energy without the shadow IT problem that comes with it.
Developers are not charged for tooling. Charges apply only to applications deployed to production.
Our Take
ServiceNow is – in our opinion – taking a savvy path that marries the ability for developers to build and deploy their own agents with ServiceNow’s proven acumen in enterprise governance. The packaging may solve the pain that IT leaders feel when they are able to budget the usage and predict outcomes. The commercial model change is real and available today, and it is a very positive but brave step.
ServiceNow already has the ability to handle Level 1 support tasks, which is something to explore. The autonomous AI story depends on Context Engine reaching general availability, and that timeline is not public. Before committing to the Prime tier, get specific answers from ServiceNow on Context Engine availability in your vertical, what token consumption looks like under full autonomous operation, and what happens operationally when an AI specialist makes a consequential error.
The platform consolidation logic is compelling. A single governed destination for AI agents, workflows, and developer output, with context that compounds over time, is exactly what most enterprises are trying to assemble from five different vendors right now. ServiceNow is betting it can deliver that whole stack in one place, and the commercial model finally makes it accessible without a procurement project to get started.
Want to Know More?
ServiceNow Q4 Results Reinforce Enterprise AI Driving Business Impact
Sense, Decide, Act, Govern: ServiceNow’s Four‑Pillar Blueprint for Enterprise AI
Create a Service Management Roadmap | Info-Tech Research Group