Serval's Catalyst AI Agent Takes Aim at ServiceNow

A $1 Billion Challenger Emerges

Serval, an AI startup now valued at roughly $1 billion, is positioning itself as a direct challenger to ServiceNow with the launch of Catalyst, an agent designed to automate enterprise IT and support workflows. According to CEO Jake Stauch, Catalyst mines historical ticket data and generates code to build automations natively, arguing that an AI-first platform can outperform legacy systems like ServiceNow that were built before the agentic era 1. The pitch is bold: rather than bolting AI features onto existing workflow software, Serval wants enterprises to rebuild their automation stack around an AI core from the ground up.

Part of a Broader Agentic Wave

Serval's move is not happening in isolation. Across the enterprise software landscape, vendors are racing to embed autonomous agents into core business operations. Zaptiva, for instance, has expanded its Agentic AI Development Services to help companies deploy autonomous "digital workforces" capable of making decisions and taking action across existing ERP and CRM systems 2. Salesforce is pushing a similar vision with its Headless 360 architecture, which aims to let agents move fluidly across CRM, ERP, and supply-chain applications without losing context as they hop between systems 6. Even outside traditional enterprise software, the trend is visible: Binance now allows AI agents built on tools like ChatGPT, Claude Code, and Cursor to execute trades on its platform, though the exchange has acknowledged that oversight of these agents still largely falls to users 3.

Governance Becomes the Central Question

As agents move from experimental pilots to production systems making real decisions, questions of control, accountability, and risk are becoming as important as raw capability. One analysis argues that identity management can no longer be treated as a back-office security function — in an agentic enterprise, identity effectively becomes the control plane that determines who or what is accountable for an action 4. Related commentary warns that the next major enterprise risk isn't the intelligence of any single agent, but the unpredictable behavior that emerges when multiple interconnected agents interact, making observability and governance across systems essential 7. Cybersecurity is also being reshaped by this shift, with autonomous agents increasingly deployed for threat detection and response, even as they introduce new categories of risk that security teams must learn to manage 8.

Real-World Deployment Is Already Underway

Beyond infrastructure providers, some companies are already putting agents into customer-facing roles. New American Funding partnered with Kastle to deploy voice AI agents offering 24/7 self-service, aiming to reduce customer hold times while maintaining regulatory compliance 5. Taken together, these developments suggest enterprises are moving quickly from experimenting with individual AI tools toward deploying networks of autonomous agents — a shift that promises efficiency gains but raises unresolved questions about trust, oversight, and who is ultimately responsible when the machines act on their own.