Auditability Makes AI Essential for Analysts
SVP of Technology & Product Strategy at FactSet on driving trust through auditability
AI assistants are becoming table stakes for analysts because the bottleneck in finance is no longer access to information, it is turning too much information into a usable answer fast enough to act. FactSet built Mercury around that exact gap, letting users ask plain language questions, pull source linked research from filings, news, and market data, generate charts, and move outputs into notes, Word, Excel, and PowerPoint instead of manually stitching together dozens of screens and documents.
-
The practical shift is from search to workflow completion. Mercury started with company research for junior bankers, then expanded into transcript analysis, portfolio commentary, and pitchbook creation, which means AI is not just finding documents, it is drafting the work product analysts are judged on.
-
Adoption in finance depends on trust, not novelty. Across FactSet, AlphaSense, and Hebbia, the winning pattern is the same, source links, audit trails, permissioning, and human review. In high stakes workflows, an answer without visible evidence is far less useful than a slightly slower answer with a trace back to the document or number behind it.
-
The competitive bar is rising quickly. AlphaSense is pushing multi threaded document research and internal knowledge search, while Hebbia goes deeper into diligence, memo writing, and agent driven workflows for finance and legal. That makes AI assistance less of a premium feature and more of a baseline expectation across the analyst tool stack.
The next step is that AI assistants become embedded in every analyst surface, not a separate chatbot. The market is moving toward systems that can pull from trusted external data, a firm’s internal notes, and workflow tools in one place, then produce auditable drafts that humans refine. Analysts who work this way will cover more companies, more documents, and more scenarios with the same headcount.