AI in Finance: Polsky Center Argues Judgment Still Holds the Pen

By Olivia Hart

The Polsky Center for Entrepreneurship and Innovation has published a piece with a blunt premise: artificial intelligence can process financial data, but it cannot replace the professional judgment that finance still demands. The article, titled "Using AI in Finance: Why Professional Judgment Still Matters," comes from the University of Chicago's entrepreneurship hub.

That framing matters because it pushes against the current tide. Vendors and consultants routinely pitch AI as an end-to-end answer for analysis, forecasting and reporting. The Polsky Center's argument lands on the other side of that ledger.

What does the piece actually claim?

The core claim is in the title, and it is a claim about division of labor. AI belongs in the toolkit. Judgment belongs to the professional. The two are not interchangeable, and treating them as such carries risk.

The piece carries weight because of where it comes from. The Polsky Center sits inside the University of Chicago's Booth School of Business ecosystem, an institution with deep ties to quantitative finance. When a Booth-affiliated center says judgment still matters, that is not a Luddite talking — it is a quant shop hedging its own bets.

What is not said?

The available text offers a thesis, not evidence. There are no case studies, no error rates, no named firms and no figures on adoption or cost in the material at hand. Readers should treat the piece as an argument from authority rather than a data-backed study, at least on the strength of what has been released.

That absence cuts both ways. It weakens the piece as proof. But it also signals that the authors expect the audience — students, founders, finance professionals — to recognize the failure modes from their own experience.

Why does this argument surface now?

Timing is doing some of the work. AI tools have moved from novelty to default in many finance functions over the past two years. The question has shifted from whether to use them to how much weight their outputs deserve.

The Polsky Center's answer, stripped to its bones: use them, but do not outsource the sign-off. A model can draft the analysis. It cannot take responsibility for it.

Where does the risk sit?

The practical takeaway for practitioners follows directly. Any AI-generated valuation, forecast or risk assessment is only as good as the human review behind it. The professional who accepts the output owns the consequences — a point that regulators and auditors have made in adjacent contexts.

Take it in plain words: the machine produces the draft, and the human signs the name.

What should readers watch next?

Expect this argument to harden into training material. Business schools, including those around the Polsky Center, are rebuilding curricula around AI literacy paired with decision accountability, and essays like this one tend to foreshadow what lands in the classroom.

The so-what is straightforward. Firms that treat AI as a junior analyst — fast, tireless, requiring supervision — will get the balance the Polsky Center describes. Firms that treat it as the decision-maker will not, and they will discover the difference when the numbers fail.