Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-2133
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-21T22:03:54.600465Z
status: published
lens: p_space
sector: business
headline: Talking Technology, Not Customers: Vocabulary Decoupling in Corporate Artificial Intelligence Disclosure
dek: Artificial intelligence (AI) is widely described as reshaping marketing administration through personalization, predictive analytics, and automated customer engagement. Whether firms describe AI as a customer-facing marketing capability or as a technology and a risk to be disclosed remains an open empirical question. This study uses mandated disclosure as an observational window on that question. Drawing on a balanced panel of 1100 firm-years from 100 U.S.-listed consumer-facing firms in 15 sectors (fiscal years…
gain_title: (none)
problem_title: Across 100 consumer-facing U.S. firms, AI language in annual reports spread widely but stayed largely disconnected from marketing vocabulary, appearing as technology and risk talk rather than customer value.
trace_subject: (none)
gain_reading: (none)
gain_evidence: (none)
problem_reading: Across 100 consumer-facing U.S. firms, AI language in annual reports spread widely but stayed largely disconnected from marketing vocabulary, appearing as technology and risk talk rather than customer value.
problem_evidence: Only 3.2% of AI sentences written in 2023 to 2025 used marketing vocabulary | AI language migrated from the Business section toward Risk Factors and MD&A
quick_read: A study of Form 10-K filings from 100 U.S.-listed consumer-facing firms over fiscal 2015 to 2025 tracked how companies talk about artificial intelligence. Disclosure became near-universal, but the language stayed mostly in technology and risk sections instead of customer-focused marketing terms.

The pattern suggests that corporate AI disclosure may not reflect how firms account for value delivered to customers, which matters for investors and researchers who read filings as signals of strategy. The author stresses that the results describe disclosure and not deployed capability, so actual marketing use of AI remains unmeasured.
limitation: The findings concern disclosure, not deployed capability, so they cannot show whether firms actually use AI in marketing.
tag: Evidence-backed problem
key_points: The share of firm-years disclosing AI rose from 28.7% in fiscal 2020 to 2022 to 91.7% in 2023 to 2025. | The intensity of AI language rose roughly twentyfold between 2021 and 2025. | Within firms, adopting AI vocabulary did not bring a meaningful rise in marketing vocabulary (standardized β = 0.02, 95% CI [−0.09, 0.13]). | Digital-commerce and financial firms disclosed AI before hospitality and travel firms.
rundown: The study used a balanced panel of 1100 firm-years from 100 U.S.-listed consumer-facing firms across 15 sectors. It measured technology and marketing vocabulary using phrase-presence indicators from the EDGAR full-text index and section-located counts from full filings.

Estimates net of common year shocks found no substantively meaningful link between AI vocabulary and marketing vocabulary. Graded and staggered-adoption estimates leave room for at most a small, lagged association.

The author calls the result vocabulary decoupling: AI entered the technology-and-risk register of annual reports largely without connection to the register used to account for customer value.
sources:
- peer_reviewed | Administrative Sciences | https://doi.org/10.3390/admsci16100518 | 2026-10-20
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

ac6c02ae58ba6cbe490260eebff93269ebd9b1ab08036c2e40a719e989c8011a

previous

0000000000000000000000000000000000000000000000000000000000000000