A new research paper explores the potential of large language models (LLMs) to identify "shadow trading" targets, a novel theory of insider trading liability involving trading in a peer firm's securities based on material nonpublic information about an economically linked company. The study applied a two-stage LLM pipeline to SEC 10-K filings to assess semantic similarity between firms and correlate it with stock returns. While the pipeline successfully identified Incyte as a close peer in the SEC v. Panuwat case, broader analysis across 30 M&A events found no significant association between semantic similarity and abnormal stock returns, questioning the empirical basis of shadow trading enforcement. AI
IMPACT This research suggests current NLP models may not be sufficient for identifying complex financial relationships relevant to insider trading enforcement.
RANK_REASON Academic paper on applying NLP/LLMs to a legal/regulatory theory. [lever_c_demoted from research: ic=1 ai=0.7]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
