Announcing Trovata AI 2.0: AI Chat, Insights and Agents for Corporate Treasury

Corporate treasury teams do not need more noise. They need faster answers, better visibility, and less manual work. That’s why we’re excited to introduce Trovata AI 2.0—the next evolution of AI inside Trovata.

AI has been part of the Trovata experience for some time. But Trovata AI 2.0 represents something bigger than an upgrade. It expands AI across the platform with a more powerful chat experience, proactive insights surfaced in context, and new AI agents designed to automate recurring treasury work. Together, these capabilities are built to reduce time-to-answer, surface more relevant intelligence, and turn repeatable treasury work into governed automation.

We’ve just unveiled Trovata AI 2.0 live, watch the full replay here.

Built for How Treasury Actually Works

Today’s treasury teams operate in environments that are complex, high-consequence, and chronically lean. Even with a TMS in place, teams still spend hours every week explaining variances, investigating activity, reconciling data, and producing executive-ready updates. 

We did not build AI for treasury for its novelty, but its leverage: reduce time-to-answer, increase frequency of insight, and automate the “glue work” that consumes teams.

Trovata AI is built directly into the Trovata, a treasury operating system, where it can work on real data and workflows inside a secure environment. That means it is not a generic chatbot layered on top, but rather, embedded intelligence tied to real treasury objects, with a governed path to automation through agents.

3 New Capabilities

Trovata AI Chat

Trovata AI Chat gives treasury teams a faster way to ask questions about their data in plain language and get useful answers instantly. It is designed to collapse time-to-answer from hours to minutes by explaining what changed, why it changed, and where to focus next. 

Common use cases include explaining changes in cash position, identifying drivers by entity, bank, or account, highlighting anomalies, and creating executive-ready narratives. Trovata AI Chat is available in both Trovata Cash and Trovata TMS.

‎

Trovata AI Insights

Trovata AI 2.0 expands intelligence beyond the prompt box.

With Trovata AI Insights, notable changes, anomalies, and patterns are surfaced automatically in context, so users can understand what matters faster the moment they log in. Instead of starting from a blank page, treasury teams can begin with relevant takeaways already surfaced inside the product.

‎

Trovata AI Agents

Trovata AI 2.0 introduces Trovata AI Agents, a new way to automate recurring treasury work.

Agents can run on schedules, triggers, or rules to monitor activity, explain variances, detect anomalies, deliver updates, and support repeatable operational workflows. Early use cases include daily cash reporting with variance explanations, anomaly detection and escalation, automated reconciliation summaries, proactive alerts, and recurring executive updates. 

Why This Corporate Treasury AI is Actually Different

What makes Trovata AI 2.0 different is not just that it does more. It is built on a whole new foundation.

Trovata AI 2.0 is built on Model Context Protocol, or MCP, as foundational infrastructure. In simple terms, MCP is the layer that connects the AI model to the right treasury context and the right approved tools inside Trovata, so answers are more grounded, more accurate, and able to support real agent workflows.

That matters because treasury AI is only as useful as the context behind it. Hallucinations happen when models do not have the data, structure, or tools they need to solve the problem correctly. MCP reduces that by design by connecting the model to the right data and the right tools for the task. Put simply, it acts like a bridge between the model and the treasury environment it needs to understand.

It also lays the groundwork for agentic workflows. A data-analysis agent needs access to treasury data and analysis tools. A payments agent needs access to the right payment tools. A treasury best-practices agent may need access to customer-specific documentation and operating context. MCP is what makes those workflows practical rather than just impressive in a demo.

Just as importantly, this architecture is durable. Because the context and tooling layer sits underneath the model, Trovata can continue evolving the model layer over time without rebuilding the entire foundation. That gives Trovata more flexibility as the AI landscape changes and creates a more extensible platform for future AI experiences.

Built With Enterprise Trust in Mind

Treasury teams operate in high-trust environments, and AI has to meet that standard.

Trovata AI operates inside a fully airgapped environment within Trovata’s AWS architecture, with no public internet access. It respects role-based access controls aligned to the TMS, keeps data encrypted in transit and at rest, and supports auditability through agent run history, logs, and traceability back to underlying treasury records.

Customer data is used to generate answers and run workflows inside the environment, not to train a shared public model. That distinction matters for finance and treasury leaders evaluating whether they can adopt AI without compromising control or trust.

The Future of Treasury, Unlocked

The bigger story behind Trovata AI 2.0 is not just better chat. It is a broader shift in how treasury work gets done.

You can ask questions in plain English. You can receive insights before you ask. And now, you can automate recurring work with agents.

Ready to see Trovata AI 2.0 in action? Book a demo today or watch the livestream recording.