The New York‑based data‑infrastructure specialist Daloopa announced on June 25, 2026 that it has built a Model Context Protocol (MCP) connector for Microsoft 365 Copilot. The integration delivers source‑verified, structured financial data directly into the familiar Microsoft Office suite, allowing analysts and portfolio managers to generate reports, build models and draft presentations without leaving Word, Excel or PowerPoint. By feeding Copilot with data that is both comprehensive—covering more than 5,500 public companies worldwide—and auditable, Daloopa aims to eliminate the “hallucination” risk that has plagued large language model applications in the investment arena.
A data‑first approach to AI in finance
Financial institutions have been quick to experiment with generative AI, yet the technology’s usefulness hinges on the quality of the underlying data. “Accuracy and trust are paramount,” Daloopa’s CEO Thomas Li explained, underscoring the firm’s long‑standing emphasis on source‑linked datapoints. Each figure in Daloopa’s repository is hyperlinked to its original filing or market source, providing a built‑in audit trail that satisfies both internal compliance teams and external regulators.
The company’s platform already supports a breadth of asset classes and market segments, delivering granular fundamentals, earnings estimates and corporate actions for a global universe of listed firms. By centralising this information in a format that LLMs can ingest without extensive preprocessing, Daloopa positions itself as a critical piece of the AI stack for hedge funds, asset managers and boutique research outfits.
How the MCP connector works with Microsoft 365 Copilot
Microsoft’s Copilot suite, embedded across Office applications, leverages large language models to turn natural‑language prompts into actionable content. Daloopa’s MCP connector acts as a bridge, feeding the Copilot engine a curated stream of verified financial data in real time. When a user asks Copilot to “draft a Q2 earnings summary for Company X” or “populate a valuation model with the latest revenue figures,” the request is routed through the MCP interface, which retrieves the relevant datapoints from Daloopa’s database and returns them in a structured format that Copilot can directly embed.
The integration is designed to be “LLM‑agnostic,” meaning it can serve not only Microsoft’s proprietary model but also third‑party engines such as Anthropic’s Claude and OpenAI’s ChatGPT, provided they adhere to the MCP standard. This flexibility broadens the connector’s appeal, allowing firms that have already adopted non‑Microsoft AI tools to benefit from Daloopa’s data without re‑architecting their pipelines.
Technical depth: LLM‑agnostic, source‑linked, and audit‑ready
At the heart of the connector lies the Model Context Protocol, an emerging industry standard that defines how external data sources can be queried and returned to LLMs in a consistent, machine‑readable schema. By conforming to MCP, Daloopa ensures that each data call includes metadata about the source, timestamp and any applicable licensing constraints.
The protocol also supports incremental updates, meaning analysts can request only the latest changes—such as a newly filed 10‑K amendment—rather than pulling an entire dataset each time. This reduces latency and bandwidth consumption, a crucial consideration for real‑time trading desks that need near‑instantaneous insights.
Business impact: From data‑driven research to faster decision‑making
The practical implications for investment professionals are significant. Traditionally, analysts spend a substantial portion of their day gathering numbers from disparate portals, reconciling inconsistencies, and manually inserting figures into spreadsheets. With Daloopa’s connector, those steps are compressed into a single natural‑language prompt, freeing up resources for higher‑order analysis.
“Using Microsoft 365 Copilot brings this data directly into the tools investors already rely on, so AI outputs are grounded in accurate, source‑linked financials rather than web‑scraped information,” Li said. The result is a workflow that blends the speed of generative AI with the rigor of verified data, potentially shortening the research cycle from days to hours.
For hedge funds that rely on rapid detection of quarterly earnings inflections, the ability to ask Copilot for a “list of companies that beat consensus revenue by more than 5 % this quarter” and receive an instantly generated, source‑cited report could translate into a measurable competitive edge.
Competitive positioning: Differentiating through data integrity
While several fintech vendors have announced AI‑enhanced analytics tools, few have tackled the data‑quality problem head‑on. Competitors often rely on web‑crawled datasets that lack provenance, exposing users to the risk of model hallucinations—a scenario where the AI fabricates information that appears plausible but is factually incorrect.
Daloopa’s insistence on source‑linked datapoints sets it apart. By providing a verifiable audit trail, the platform aligns with the heightened compliance expectations of the financial sector, where regulators such as the SEC are increasingly scrutinising the use of AI in investment decision‑making. The company’s existing partnerships with OpenAI and Anthropic further reinforce its credibility as a neutral data provider that can serve multiple AI ecosystems.
Industry context: AI adoption and regulatory scrutiny
The broader financial services industry is in the midst of an AI renaissance. According to a recent survey by the CFA Institute, over 70 % of asset‑management firms have piloted generative‑AI tools, yet concerns about data fidelity and model transparency persist. Regulators in the U.S. and Europe have signalled that firms must retain the ability to explain AI‑driven recommendations, a requirement that dovetails with Daloopa’s source‑linked architecture.
Moreover, the rise of “agentic” workflows—where autonomous AI agents execute tasks such as trade execution or risk monitoring—demands a reliable data backbone. Daloopa’s MCP connector could serve as a foundational layer for these agents, ensuring that autonomous decisions are rooted in verifiable market information.
Future outlook: Scaling the connector and expanding use cases
Daloopa’s roadmap indicates that the MCP integration with Microsoft 365 Copilot is just the first step. The company plans to extend the connector to additional Microsoft products, including Teams and Power BI, and to deepen its support for other LLM providers as the MCP ecosystem matures. In parallel, Daloopa is exploring the incorporation of alternative‑data streams—such as ESG scores and satellite‑derived foot‑traffic metrics—into its structured offering.
If the adoption curve follows that of other AI‑enhanced financial tools, the connector could become a de‑facto standard for embedding high‑quality data into generative‑AI workflows across the sector. The key will be maintaining the balance between openness (supporting multiple AI models) and data governance (ensuring compliance and auditability).
Conclusion
Daloopa’s Model Context Protocol connector for Microsoft 365 Copilot represents a pragmatic response to the twin challenges of AI adoption in finance: speed and data integrity. By delivering source‑verified financial information directly into the everyday tools of analysts, the integration promises to reduce manual effort, improve model reliability and align AI‑driven insights with regulatory expectations. As the financial industry continues to embed generative AI into its core processes, solutions that marry robust data infrastructure with flexible AI interfaces—like Daloopa’s—are likely to shape the next wave of productivity gains.
For more details or to request a demonstration, visit daloopa.com.
Get in touch with our fintech experts
