The SEC has cautioned public firms against artificial intelligence (AI) washing—overstating AI investments in corporate disclosures. Securities law experts expect AI washing to increase as firms face intensified competitive pressures to deploy AI, but disclosure theory suggests competition may induce AI hushing—understating AI investments in disclosures—due to proprietary cost concerns. I examine whether technological peer pressure (TPP) fuels AI washing or hushing. Using a word embedding machine learning model, I construct AI and investment dictionaries, and measure AI washing or hushing as the difference between a firm’s decile rank in retrospective AI investment disclosure and its decile rank in actual AI investment among peer firms in a year. To improve identification, I exploit state R&D tax credits and AI innovation breakthroughs to capture plausibly exogenous increases in TPP. I document that TPP induces AI washing, particularly among firms opportunistically overstating AI investment, benefiting more from capital market rewards, or seeking strategic competitive advantages. However, the effect vanishes when proprietary cost concerns are high. TPP-induced AI washing enables washers to attract more capital but leaves them with weaker operational efficiency and marginally lower AI innovation than firms reporting AI investment accurately. Overall, my findings reveal how technological competition distorts AI investment reporting, raising pressing regulatory concerns given the SEC’s mandate to ensure truthful disclosures that facilitate efficient capital allocation.

wlareina@ceibs.edu