Humans on Tuesdaylaunched a suite of AI toolsTargeted directly at the financial industry, the product introduced 10 pre-built agent templates designed to automate labor-intensive tasks in banking, asset management, and insurance.
These templates—provided as plugins in Claude Cowork and Claude Code, and as “recipes” in Claude Managed Agents—are designed to streamline tasks that have long consumed significant time from analysts, including creating pitch materials, screening client due diligence documents, and running month-end closing processes. The company says businesses can have Claude handle real financial tasks in just days, not months.
These agents are divided into two main categories. In terms of research and client coverage, the templates include: outreach tools for creating target client lists and drafting pitch materials; meeting preparation tools for organizing client briefs; and profit review tools for reading meeting transcripts and flagging changes related to arguments.
Another group of finance and operations agents handles tasks such as general ledger reconciliation, month-end closing lists, financial statement audits, and KYC compliance screenings.
Each template consists of three components: skills, domain knowledge for coding tasks; connectors, for controlled access to relevant data; and sub-agents, or additional Claude models for specific subtasks such as comparable object selection or method verification.
This release also expands Claude’s integration within the Microsoft Office suite. Claude can now work directly across Excel, PowerPoint, Word, and Outlook—context is automatically synchronized between applications, so analysts no longer need to re-explain model content in PowerPoint after creating financial models in Excel.
Anthropic is also expanding its data partner ecosystem with new connectors from Dun & Bradstreet, Fiscal AI, Guidepoint, IBISWorld, SS&C IntraLinks, Third Bridge, and Verisk, as well as the MCP application from Moody’s, which provides proprietary credit ratings and data for over 600 million public and private companies.
Combined with the new feature and Claude Opus 4.7, Anthropic states that this version leads the industry with a score of 64.37% on the Vals AI financial agent benchmark. Anthropic also notes that users can review and approve Claude’s outputs before they are delivered to clients or archived.
Anthropic recently garnered significant attention with its Claude Mythos model, marking a turning point in artificial intelligence’s ability to discover and potentially exploit software vulnerabilities. The company is currently making this model available only to select users.Selected tech companies and institutionsSo far, Mozilla has reported discovering andfixing hundreds of vulnerabilitiesthanks to this tool, which is integrated into the Firefox browser.
On Tuesday, Dario Amodei, co-founder and CEO of Anthropic, issued a warning.“A moment of danger”The tech and artificial intelligence industries must prepare thoroughly and act cautiously to prevent attackers from gaining access to powerful AI capable of disrupting critical software and systems.
Of course, this is not the only concern brought about by artificial intelligence. The tools launched on Tuesday to alleviate tedious work in the financial industry highlight ongoing fears that AI could disrupt industries across the board. Leading to mass layoffs.
Companies like this yuan,clogged andCoinbase layoffs are tied to plans to boost efficiency through artificial intelligence, and these predictions are quickly becoming reality.