% ============================================================================
\section{Related Literature and Hypotheses}
\label{sec:literature}
% ============================================================================

Our study sits at the intersection of five strands of research: textual analysis in finance, the
pricing of corporate disclosure, technology and innovation in asset pricing, investor attention and
narrative economics, and the nascent literature on the economics of artificial intelligence. Each
contributes a building block---a measurement tool, an object of study, a candidate risk factor, a
behavioural mechanism, and a comparison case---and a sixth, methodological strand disciplines the
inference. We take the strands in turn and close with the hypotheses they jointly motivate.

\subsection{Textual analysis and natural language processing in finance}
A now-large literature extracts economic signals from financial text. \citet{tetlock2007} showed
that the tone of a \textit{Wall Street Journal} column forecasts aggregate returns, and
\citet{tetlock2008} extended this to firm-level news. \citet{lm2011} demonstrated that
finance-specific sentiment dictionaries materially outperform general-purpose word lists in 10-K
analysis, launching a wave of dictionary-based studies; \citet{lm2016} and \citet{lm2020} survey
the field and its pitfalls, including the sensitivity of results to document selection and
pre-processing. A second generation moved from counting words to modelling them: \citet{gentzkow2019}
review text-as-data methods in economics, \citet{kekellyxiu2019} develop a supervised topic model
that predicts returns from news, \citet{kelly2019textregression} propose a sparse text-regression
framework, and \citet{bybee2024narrative} recover a ``state of the economy'' from business news.
A third generation uses distributed representations of meaning. \citet{hanleyhoberg2019} apply word
embeddings to bank 10-Ks to identify emerging risks before they are named in dictionaries;
\citet{limai2021} measure corporate culture from earnings calls with word embeddings;
\citet{huang2023finbert} show that a domain-adapted transformer outperforms dictionaries and
bag-of-words classifiers on financial text; and \citet{cong2022textfactor} build interpretable
``textual factors'' from embeddings. Our paper belongs to this embedding-based generation. The
limitations of keyword counting are acute for a theme like AI: a count cannot distinguish ``we do
not currently use artificial intelligence'' from ``we are an artificial-intelligence leader,'' it
misses the paraphrase of a firm that describes machine learning without writing ``AI,'' and it
treats boilerplate and substance identically. We instead measure the semantic similarity of each
passage to anchor concepts with a sentence transformer \citep{reimers2019sbert}, a design close in
spirit to the similarity-to-reference-text constructions of \citet{hanleyhoberg2019}. The
literature also warns, however, that text-based measures inherit the composition of the corpus they
are computed on \citep{lm2016,lm2020}; Section~\ref{sec:data} therefore documents that composition
in detail.

\subsection{The pricing of corporate disclosure}
A parallel literature asks whether the \emph{content} and \emph{changes} of mandatory disclosure
predict returns. \citet{cohen2020lazy} show that firms that change the language of their annual
reports relative to the prior year experience predictable future returns and operating outcomes.
\citet{hoberg2010product} and \citet{hobergphillips2016} use 10-K product descriptions to build
text-based industry classifications that carry asset-pricing content. \citet{lopezlira2023risk}
extracts priced risk exposures from risk-factor sections. The general lesson is that the
qualitative portions of filings contain value-relevant information that markets process
incompletely and with delay. We contribute a concept-level measure---AI engagement---computed
mainly from current reports rather than annual reports, and we characterize how its relation to
returns varies over time, a dimension the disclosure-pricing literature has largely left implicit.

\subsection{Technology, innovation, and the cross-section of returns}
Asset pricing has long sought to connect technological change to risk and return.
\citet{hirshleifer2013innovative} show that innovative efficiency predicts returns;
\citet{kogan2017technological} build market-based measures of innovation from patents and link them
to growth and reallocation; \citet{koganpapanikolaou2014} show that investment-specific technology
shocks shape the cross-section. A more recent literature studies AI exposure directly.
\citet{babina2024aiinvestment} measure firm-level AI investment from job postings and r\'esum\'es
and document faster growth among AI-adopting firms; \citet{acemoglu2022aijobs} and \citet{webb2020}
measure the exposure of occupations and establishments to AI; \citet{acemoglu2020wrong} study the
labour-market effects of industrial robots. Closest to us, \citet{eisfeldt2023generative} construct
a firm-level measure of \emph{generative}-AI exposure from the task content of occupations and show
that a long--short portfolio on that measure earned large returns in the weeks after ChatGPT's
release; \citet{jha2024chatgpt} use ChatGPT itself to extract expected investment and employment
policies from earnings calls; and \citet{lopezliratang2026} show that a large language model can
rank news headlines in a way that predicts next-day returns. Our contribution relative to this work
is to measure AI exposure from the firm's own disclosure language, available for the entire index
at annual frequency, and to characterize the full time profile of its pricing across fifteen
cohorts rather than an event window. Where \citet{eisfeldt2023generative} document a sharp
repricing in the weeks around the launch, we ask whether a persistent, text-based exposure measure
carried a premium before the launch and whether the premium after it is distinguishable from a
technology-sector effect.

\subsection{Investor attention, salience, and narrative economics}
Why might value-relevant disclosure be priced only after a catalyst? The behavioural literature
offers a mechanism: limited investor attention. \citet{barberodean2008} show that individual
investors are drawn to attention-grabbing stocks; \citet{da2011} construct an attention proxy from
Google search and link it to prices; \citet{dellavigna2009} document under-reaction to information
released when attention is low; and \citet{hirshleifer2009driven} show that earnings news is
underweighted when investors are distracted. \citet{shiller2017narrative} argues that contagious
narratives drive major asset-price movements, and \citet{bakerwurgler2006} show that sentiment
differentially affects hard-to-value, speculative stocks. These mechanisms predict that a salient
consumer event such as ChatGPT could draw attention to information that had been sitting in filings
for years. They do not, however, distinguish a firm-level attention shock from a sector-level
revision of expected cash flows, and our tests are designed to separate the two.

\subsection{ESG and climate disclosure as a comparison case}
Our placebo design connects to the literature on the pricing of ESG and climate disclosure.
\citet{pastor2021sustainable} model how sustainability preferences depress the expected returns of
green assets in equilibrium, and \citet{pastor2022dissecting} show that green stocks' strong
realized returns through 2021 reflected unexpected increases in climate concern rather than high
expected returns. \citet{bolton2021risk} document a carbon premium, and \citet{sautner2023} build a
firm-level climate-exposure measure from earnings-call language that is priced in options and
returns. Consistent with the post-2021 reversal in this literature, we find that high-ESG and
high-climate disclosure earned negative spreads after 2022, which helps to rule out a generic
``more disclosure earns more'' explanation of the AI result.

\subsection{Short samples, data snooping, and event-study inference}
Two methodological literatures bear directly on how much weight our post-2022 estimates can carry.
The first concerns the credibility of cross-sectional return patterns. \citet{harveyliuzhu2016}
argue that the multiplicity of tested predictors requires $t$-statistics well above conventional
thresholds; \citet{houxuezhang2020} find that most published anomalies fail to replicate under
standard procedures; and \citet{mcleanpontiff2016} document post-publication decay. A spread
estimated on three annual observations, one of which is extreme, must be read in that light. The
second concerns inference in event studies when all firms share a single event date, so that
abnormal returns are cross-sectionally correlated \citep{brownwarner1985,kolari2010} and
event-induced variance may differ from estimation-window variance \citep{boehmer1991}, and in
long-horizon settings, where the choice of benchmark and the compounding of abnormal returns
matter \citep{barberlyon1997,fama1998,mitchellstafford2000,kothariwarner2007}. Our event-study
inference follows the portfolio approach recommended in that literature. Finally, the use of a
current-membership universe raises the survivorship concerns analysed by
\citet{browngoetzmannross1995}, which we address with restrictions on listing history.

\subsection{Hypotheses}
\label{subsec:hypotheses}
Our analysis is organized around four hypotheses.
\begin{description}[leftmargin=2.4em,style=nextline]
\item[H1 (Unconditional pricing).] If AI exposure carries a stable risk premium or persistent
mispricing, then sorting on AI-disclosure intensity yields a positive and significant risk-adjusted
return spread over the full 2010--2024 formation sample.
\item[H2 (Regime dependence).] If instead AI exposure is repriced only once AI becomes salient, the
return spread is small before and large after the late-2022 generative-AI catalyst, and the
increase is visible in each post-launch cohort rather than in a single one.
\item[H3 (AI-specificity).] If the post-2022 spread reflects an AI repricing rather than a
mechanical or market-wide artefact, it appears for AI-related semantic factors but not for unrelated
factors such as ESG, climate, or macro risk, and it is not spanned by a technology-sector factor.
\item[H4 (Cross-sectional, not within-firm).] If the market repriced \emph{which} firms are
AI-exposed rather than rewarding incremental wording, the effect is concentrated in the
cross-section, attenuates under firm fixed effects, and displays no pre-trend when exposure is held
fixed at its pre-launch level.
\end{description}
\noindent The evidence rejects H1, supports the first clause of H2 but not the second, supports
H3 with respect to the placebo factors but not with respect to the sector factor, and supports H4
only in part.