Stockholm-based startup Vesper has emerged from stealth mode with an AI-powered platform designed to transform how finance teams review, control and explain financial statements.
The company is building an agentic workflow for financial statement control, bringing several traditionally fragmented controller tasks into a single workspace. Vesper’s platform is designed to help finance teams review P&L and balance sheet statements, investigate account and transaction-level issues, identify the drivers behind financial movements and create management-ready explanations.
The startup was first reported by Swedish Tech News in June and has now publicly unveiled its product.
Bringing financial control into one workflow
Financial statement review often requires controllers and accountants to move between consolidation software, accounting systems, ERP platforms, spreadsheets and presentation tools.
Vesper aims to replace this fragmented process with a connected workflow.
Its platform allows finance teams to:
- Review P&L and balance sheet statements
- Run configurable account and transaction controls
- Drill down from financial statement lines to individual GL transactions
- Create and manage exception and fix lists
- Investigate why financial figures changed
- Generate CFO- and FP&A-ready narratives
According to Vesper, the platform can work directly from files that controllers already export from their accounting systems, allowing teams to get started without an ERP integration or lengthy implementation process.
AI grounded in financial data
A key component of Vesper’s offering is Whisper AI, which allows users to ask questions about their general ledger and financial statements.
Instead of providing an answer without context, the system combines the statement, account, issue and transaction information before generating a response. This is intended to make AI-generated explanations traceable to the underlying financial data.
For example, a controller can investigate why revenue changed, identify which transactions drove an account movement or compare changes against the previous reporting period.
This approach reflects a broader shift in enterprise AI toward agentic workflows, where AI is embedded into specific business processes rather than functioning solely as a general-purpose chatbot.
Designed for sensitive financial data
Because Vesper works with financial statements and general ledger information, security and data governance are central to the platform.
The company says customer data is encrypted both in transit and at rest, workspaces are isolated by organization and user permissions, and customer data is not used for model training unless explicitly agreed to or opted into. The platform is hosted in the EU by default, according to Vesper.
Vesper also maintains a traceable record of actions through what it calls Vesper Vault, giving finance teams an audit trail of the review process.
Targeting the modern finance function
The emergence of Vesper comes as finance departments face growing pressure to close faster, improve financial controls and provide executives with clearer explanations of business performance.
Traditional accounting and ERP systems contain the underlying financial data, but the review process often still depends heavily on spreadsheets, manual investigation and conversations between finance teams.
Vesper is positioning itself between these systems and the people responsible for interpreting the numbers—using AI to connect financial data, deterministic controls and the controller’s workflow.
The startup’s broader ambition is to make financial statement control more continuous, structured and explainable, while giving finance professionals more time to focus on analysis and decision-making rather than manually searching for discrepancies.
With its emergence from stealth, Vesper joins a growing wave of Swedish startups applying AI agents to highly specialized enterprise workflows—this time targeting one of the most critical functions inside every business: understanding and controlling its financial statements.
