Between 2025 and 2026, ServiceNow consolidated its agentic AI strategy with multi-model partnerships (OpenAI and Anthropic), the acquisitions of Moveworks (~US$ 3 billion), Armis (US$ 7.75 billion), Veza and Pyramid Analytics, a revamped partner program and, in Knowledge 2026, the launch of Autonomous Security & Risk — positioning the platform as an “AI control tower” for more than 100 billion workflows per year.
In less than twelve months, ServiceNow transformed its positioning from a workflow platform into an execution infrastructure for agentic AI. This involved agreements with leading AI labs, a series of acquisitions focused on data, identity, and security, a redesign of its partner program, and, at Knowledge 2026, the announcement of autonomous governance capabilities. This article consolidates the most relevant moves and what they mean for those operating ServiceNow in Brazil.
Timeline of agentic AI in ServiceNow
| Period | Movement | Essence |
|---|---|---|
| March 2025 | Acquisition of Moveworks | ~US$ 3 billion; corporate search and conversational AI for employee support |
| Dec/2025 | Veza and Armis's Acquisitions | AI-native identity security (Veza) and asset intelligence (Armis, US$ 7.75 billion) |
| 20/01/2026 | Partnership with OpenAI | GPT-5.2 as a preferred intelligence capability; speech-to-speech voice. |
| 20/01/2026 | Revamped Partner Program | Build Program opens and ServiceNow Store serves as a marketplace for AI agents. |
| 28/01/2026 | Partnership with Anthropic | Claude as the standard ServiceNow Build Agent template |
| Feb/2026 | Acquisition of Pyramid Analytics | Analytics and BI with natural language querying (closed on 10/03/2026) |
| May 2026 | Knowledge 2026 | Launch of Autonomous Security & Risk, integrating Armis and Veza. |
Multi-model partnerships: OpenAI and Anthropic
ServiceNow has explicitly adopted a multi-model strategy—the stated goal is to offer the right model for each task while maintaining consistent governance, security, and auditability across the platform. This agentic layer operates alongside the Now Assist and generative AI native to ServiceNow.
ServiceNow and OpenAI
Announced on January 20, 2026, the multi-year agreement This makes OpenAI models a preferred intelligence capability for the more than 80 billion workflows that run annually on the platform. The integration of GPT-5.2 into the ServiceNow AI Platform allows AI to understand context, decide the next step, and execute actions within the customer's governed infrastructure. The agreement also provides for speech-to-speech technology and computer use models for automating legacy systems.
ServiceNow and Anthropic
One week later, on January 28, 2026, ServiceNow signed an agreement with Anthropic to make Claude the standard ServiceNow Build Agent model and the preferred model in application development workflows. The company reports having deployed Claude to its more than 29,000 employees—reducing sales preparation time by up to 95%—and made Claude Code available to engineering teams. The collaboration targets agency solutions for regulated sectors, such as healthcare and life sciences, where claims authorization could drop from days to hours.
Acquisitions: data, identity, and security
The acquisition strategy reveals ServiceNow's thesis: for AI agents to act with confidence, the platform needs to master data, identity, and security end-to-end.
- Moveworks (~US$ 3 billion, 2025): corporate search and conversational AI, expanding its reach beyond IT — HR and customer service.
- Armis (US$ 7.75 billion): continuous asset intelligence in IT, OT, IoT and connected devices, directly feeding the CMDB.
- Veza (Closed on 02/03/2026): AI-native identity security, whose Access Graph maps access relationships of human, machine, and AI agent identities.
- Pyramid Analytics (Closed on 10/03/2026): Analytics, data science, and BI platform with natural language querying, adding a semantic layer that humans and agents can use to act.
Autonomous Workforce: From Assistive AI to Autonomous AI
Based on these pieces of evidence, ServiceNow structured the concept of Autonomous Workforce: AI specialists assigned to roles, with business context and permissions to conduct complex end-to-end workflows. EmployeeWorks is already available to customers, and Level 1 Service Desk AI Specialist is headed for broad availability in the second quarter of 2026. This move marks the transition from AI that merely assists to AI that reliably performs work under governance.
Partner Program reimagined
At its Partner Kickoff on January 20, 2026, ServiceNow revamped its global partner program to accelerate innovation in AI agents. The new Build Program opens the ecosystem to more ISVs and developers and transforms the ServiceNow Store into a global marketplace for partner-built AI agents, with over 1,000 partners migrating to the revised model within an ecosystem of over 2,700 partners. The structure now features tiers (Registered, Select, Premier, and Elite), simplified pricing, and a unified investment portfolio (MDF, SIF, and sales and specialization incentives).
Highlights of Knowledge 2026
At Knowledge 2026 in Las Vegas, ServiceNow connected its acquisitions and partnerships to concrete launches:
- Autonomous Security & Risk: integrates Armis and Veza into a single platform for governing AI agent identities, permissions, and connected assets.
- New AI Specialists for vulnerability resolution and security operations, acting autonomously.
- AI Control Tower Expanded, with governance extended to more than 30 systems, including cloud providers and AI models, as well as integration with Microsoft Agent 365.
- Programs with Accenture (forward deployed engineering), AWS and NVIDIA (Project Arc) to take pilot agents to production scale.
What does this mean for the Brazilian market?
For Brazilian organizations, the message is practical: the window for experimentation with AI is closing, and the phase of large-scale implementation has begun. This scenario connects to... ServiceNow trends for 2026. Three implications stand out:
- Fundamentals come first. Trusted AI agents rely on contextualized data (strong CMDB and CSDM), governed identity, and security—precisely the areas strengthened by the acquisitions.
- Multimodeling requires governance. With OpenAI and Anthropic on the platform, the choice of model becomes an architectural and compliance decision, not just a product decision.
- Specialized partners accelerate adoption. The new partner program and the focus on AI-assisted delivery make partner maturity a decisive factor in speed and risk.
As a ServiceNow Elite Partner, 4MATT tracks these trends to help companies transform data into action with governance — from maturity diagnosis to agency operations. ITSM, ITOM, CMDB and asset management.
Frequently Asked Questions
What is ServiceNow's AI strategy?
ServiceNow adopts a multi-model and agentic approach: it integrates OpenAI and Anthropic models into its platform, strengthens data, identity, and security through acquisitions, and governs everything in an "AI control tower".
Which companies has ServiceNow acquired recently?
Between 2025 and 2026, ServiceNow acquired Moveworks (search and conversational AI), Armis (asset intelligence), Veza (identity security), and Pyramid Analytics (analytics and BI).
What is Autonomous Security & Risk?
It is the solution launched at Knowledge 2026 that integrates Armis and Veza to govern AI agent identities, permissions, and connected assets on a single platform.
Does ServiceNow use OpenAI or Anthropic?
Both. OpenAI's GPT-5.2 is the preferred intelligence capability on the platform, and Anthropic's Claude is the default ServiceNow Build Agent model—part of a multi-model strategy.
About 4MATT
4MATT Technology is ServiceNow Elite Partner, with approximately 138 employees and 118 ServiceNow certified professionals. Recognized with the ServiceNow Technology Excellence Partner Award (2024 and 2025), is an expert in ITAM, ITOM, CMDB, ITSM, SAM, and IT governance, supporting organizations in adopting agentic AI with trusted data and end-to-end governance.