



The Andersen Institute partnered with Andersen Consulting’s AI & Advanced Analytics practice to develop tools to better understand how tariff increases are affecting corporate behavior and profitability.
Using Palantir Foundry and large language models, Andersen built an AI-enabled economic intelligence platform.
Within weeks, the Institute gained a market-wide view tariff impacts and mitigations strategies.
The Andersen Institute partnered with Andersen Consulting’s AI & Advanced Analytics practice to understand how the largest U.S. tariff increase since the Great Depression was affecting corporate behavior and profitability. Using Palantir Foundry and frontier large language models (LLMs), researchers at the Andersen Institute built an AI-enabled economic intelligence platform capable of analyzing nearly 1,500 earnings call transcripts for S&P 500 firms over the first three quarters of 2025 and converting qualitative executive commentary into structured, measurable insights.
Within weeks, the Institute gained a market-wide view of tariff impacts and mitigation strategies across individual firms, sectors, and time, something previously impossible with manual research methods. The platform revealed consistent patterns across sectors, linked executive statements to realized margin outcomes, and established a scalable foundation for ongoing, data-driven economic analysis.
Across global financial markets, shifts in trade policy have become major drivers of corporate performance. When the U.S. enacted a historic tariff increase in early 2025, research teams faced immediate pressure to understand how companies were responding in terms of costs, supply chains, pricing, productivity, and investment decisions.
Yet traditional economic analysis tools were not built for this moment. Traditionally, researchers have relied on manual review of individual transcripts, fragmented datasets, and inconsistent interpretations of executive commentary. Capturing insights across the entire S&P 500 in a matter of weeks, let alone with consistency and rigor, was unattainable.
For the Institute, the challenge was especially complex:
The Institute needed a modern research environment—one capable of fusing heterogeneous data, extracting structured signals from unstructured text, and producing timely, repeatable insights that senior decision makers could trust.
To meet this challenge, researchers at the Andersen Institute worked with Andersen Consulting to assemble a cross-functional AI and economics analytics team to design and deploy a next-generation research capability. Built on Palantir Foundry and powered by advanced LLMs, the platform automatically processes earnings call transcripts, identifies tariff-related commentary, and extracts standardized, causal indicators of business impacts and mitigation strategies.
The result: a repeatable system that transforms qualitative corporate dialogue into clean, economic insight at national scale.
Before this initiative, analysts pulled information from multiple disconnected systems like raw transcripts, sector classifications, financial statements, and quarterly margin data. These sources had different structures, formats, and update cycles, slowing analysis and undermining comparability.
Foundry unified these assets into a single, governed research environment. All transcripts, metadata, and financial results flowed into a consistent data model, giving analysts a reliable basis for cross-sectional and time-series analysis. With this foundation, researchers at the Institute were able to examine qualitative and quantitative information in a single research context.
AI That Turns Qualitative Commentary into Structured Economic Signals
The Andersen Institute utilized a two-stage LLM classification pipeline to interpret each transcript with precision:
Each transcript produced 39 structured variables, enabling consistent interpretation across all S&P 500 firms. This approach eliminated months of manual reading and coding, replacing it with a deterministic, fully auditable AI workflow.
To empower economists and analysts, the Andersen Institute working with Andersen Consulting built intuitive dashboards directly inside Foundry. These dashboards provided:
Researchers gained immediate access to insights that previously required multiple tools and manual interpretation.
The system was implemented quickly and seamlessly integrated into the Institute’s research tools. New earnings call transcripts are now automatically ingested, processed, and added to the earnings transcript database.
Leadership benefits from a continuous, consistent view of evolving corporate sentiment and economic pressures—without additional manual workload.
The collaboration delivered measurable value across analytical, operational, and strategic dimensions:
This tool gives the Institute a faster way to measure how markets feel about economic conditions, check those signals against what actually happens, and deliver sharper, more grounded advice.
Through its partnership with Andersen Consulting’s AI & Advanced Analytics practice, the Andersen Institute has demonstrated how AI and human expertise can work together to elevate economic insight. Complex, unstructured commentary—once impossible to analyze at scale—now serves as a strategic asset.
“This platform gives our research teams the ability to interpret the corporate sector at scale,” said, Alessandro Rebucci, the Institute’s Head of Research. “It has fundamentally changed how we assess economic shocks and communicate insights to leadership.”
Looking ahead, the Institute plans to extend the platform to incorporate real-time regulatory filings, global earnings data, alternative datasets, and predictive analytics—setting the stage for a fully AI-enabled research organization capable of responding to geopolitical and macroeconomic developments with unprecedented speed and precision.
©2025 Andersen Institute for Finance & Economics. All Rights Reserved. This material is confidential intellectual property of the Andersen Institute for Finance & Economics. The views expressed in this note are those of the authors and do not represent an official position of The Andersen Institute for Finance and Economics or affiliated organizations. By viewing this Andersen Institute Insight, you agree that you will not directly or indirectly copy, modify, record, publish, or redistribute this material and the information therein, in whole or in part. No warranty or representation, express or implied, is made by the Andersen Institute or any of its affiliates, nor does Andersen accept any liability with respect to the information and data set forth herein. Distribution hereof does not constitute legal, tax, accounting, investment or other professional advice. The information provided herein is not intended to provide a sufficient basis on which to make an investment decision. Recipients should consult their own advisors, including tax advisors, before making any investment.