Anthropic has made an expansion of its Claude AI platform for financial services, unveiling a suite of new tools designed to bring real-time intelligence and automation to the desks of analysts, bankers, and asset managers.
The update introduces an Excel add-in, additional market data connectors, and new pre-built Agent Skills for complex financial modeling and reporting tasks.
These enhancements build on Claude’s recent financial reasoning. The Sonnet 4.5 model achieved 55.3% accuracy on the Finance Agent benchmark by Vals AI, outperforming other systems in tasks such as portfolio analysis and equity valuation. Together, these upgrades signal Anthropic’s growing ambition to make Claude indispensable across the financial sector.
AI embedded directly in spreadsheets
Claude for Excel is now in beta. The feature embeds Claude into Microsoft Excel as a sidebar, allowing users to chat with the AI while it reads, analyzes, and modifies spreadsheets in real time. Claude can explain its reasoning, track every change it makes, and link users directly to the affected cells.
A typical use case involves financial analysts working with income statements or forecast models. Claude can debug formulas, preserve dependency chains, and build new sheets entirely from scratch — all while providing natural language explanations of its logic. This hands-on transparency aims to address long-standing concerns in finance about AI-driven errors in sensitive calculations.
The add-in joins Claude’s existing integrations across Microsoft 365, where the AI can already edit PowerPoint decks, draft reports, and retrieve files, emails, or Teams messages. The beta is available as a research preview for Claude Max, Enterprise, and Teams users, with 1,000 initial participants offering feedback before wider release.
Connecting Claude to live financial ecosystems
Anthropic is also expanding Claude’s access to real-time information through new data connectors, effectively transforming the model into a live market companion. Connectors from leading platforms such as Aiera, Chronograph, Egnyte, LSEG, Moody’s, and MT Newswires enable Claude to analyze current financial conditions, corporate events, and macroeconomic trends.
For example, Aiera integrates real-time earnings call transcripts and investor event summaries, while its partnership with Third Bridge gives Claude access to expert interviews and sector intelligence. Chronograph adds private equity insights, including valuation data and fund performance, whereas Moody’s provides credit ratings and data on over 600 million companies. These connections bring the immediacy of Bloomberg-style feeds into the conversational interface of Claude.
By tying Claude directly to market data, Anthropic is trying to position its AI to support financial professionals who need to synthesize fast-moving information into strategic insights. The goal is to cut down research time and reduce human error in data gathering and analysis.
New agent skills
Alongside these integrations, Anthropic has introduced six new Agent Skills designed specifically for finance. Skills are essentially task modules containing predefined scripts, data structures, and best-practice logic that Claude can use across its applications and API.
The new skills automate common workflows such as discounted cash flow modeling, comparable company analysis, and due diligence. Claude can also generate company profiles, earnings summaries, and full “initiating coverage” reports — work typically handled by junior analysts or research associates. Each output is dynamic, allowing for refreshed data or customized assumptions, and integrates seamlessly into existing financial documents and presentations.
These capabilities illustrate how generative AI is moving beyond chat interfaces into practical automation, freeing professionals to focus on higher-value judgment and strategy rather than repetitive spreadsheet work.
The broader implications for financial institutions
Claude’s growing presence in financial services reflects a broader industry shift toward AI-assisted analysis. Major banks, insurers, and fintech companies already use Claude to streamline client engagement, risk modeling, compliance documentation, and legacy system modernization.
The integration of advanced financial reasoning and direct data connectivity could reshape how analysts, underwriters, and advisors operate — compressing workflows that once took hours into minutes. By embedding AI directly within core tools like Excel and connecting it to real-time intelligence networks, Anthropic is challenging traditional financial software providers and accelerating the transition to AI-augmented finance.
Claude’s evolution underscores how generative AI is no longer an experimental aid but arguably a central infrastructure for decision-making in global finance. It will probably also lead to job cuts, as firms will employ fewer people if some of their work can be done by AI.
Anthropic recently announced a major expansion of its collaboration with Google Cloud, signaling one of the largest compute scale-ups in the history of AI infrastructure.
This article was reviewed by Antony Peyton.
