Talks, research, teaching, and tools focused on using artificial intelligence without giving up verification, judgment, or researcher ownership.
AI is rapidly lowering the cost of producing analysis, code, and prose. It is not lowering the standard for evidence. As plausible outputs become easier to produce, verification, judgment, and claim ownership become more important—and often more costly. Researchers still have to decide which questions are worth pursuing, which evidence is credible, and which claims they are willing to put their names behind.
My work in this area examines both sides of that change: how AI affects the production and communication of knowledge, and how researchers can use agentic tools while preserving transparent, reproducible workflows. The objective is not automation for its own sake. It is to use the technology where it expands careful inquiry while retaining human responsibility for the result.
AI expands the set of ideas, analyses, and claims that researchers can produce. The resulting constraint is increasingly the capacity to verify those outputs, exercise judgment over them, and take ownership of the claims that survive. This talk develops that argument and considers its implications for research design, doctoral training, collaboration, and the marketplace of ideas.
With John Campbell, Ryan Grant Johnson, and Christine Liu. We examine whether corporate AI disclosures provide credible information about firms' underlying AI investments and future economic outcomes, or primarily reflect inexpensive claims.
This hands-on PhD course focuses on agentic AI for empirical accounting research. Five modules of hands-on work incorporate AI into the research pipeline responsibly: the course starts at first install and ends at a verified WRDS-to-Stata pipeline, covering reproducible pipelines, text measures from EDGAR, licensed data on your own account, and the verification protocol that makes the speed usable in a paper.
The course closes with a seminar on the economics of knowledge when machines can produce it. For the consumer, AI-created knowledge is privately cheap but socially valuable only if it can be verified and trusted. For the producer, AI lowers the private cost of research, while the social return depends on quality, congestion, and the allocation of scholarly attention. The seminar asks where these private and social margins diverge in the academic labor market, and what remains scarce.
A free, open-source collection of 46 agent skills and a WRDS data connector for empirical researchers in economics and accounting. The skills cover econometric workflows, academic writing, document preparation, and data visualization; the connector lets agents query Compustat, CRSP, and other WRDS databases with your own account. Works with Claude, Cursor, and Codex.