Kusumegi et al. (2025) study whether researchers' preprint output rises after adopting large language models (LLMs), dating adoption as the first month in which at least one submitted abstract exceeds an LLM-detection threshold. We show that this treatment-timing rule is mechanically related to output. The probability that at least one paper is flagged in a month is increasing in the number of papers submitted in that month, so detected-adoption months are disproportionately high-output months. An event study centered on first detection can therefore display positive post-event dynamics even when the flagging rule contains no information about true LLM adoption, because the omitted pre-treatment period is selected from months with no prior detection. We demonstrate this in a simulation: with i.i.d. productivity and no causal effect, first-detection timing generates a spurious positive post-treatment path. We also replicate the stacked event study of Kusumegi et al. (2025) and show that three placebo exercises (random paper-level assignment, neutral keyword flags, and a pre-ChatGPT observation window) each produce a similarly positive post-treatment pattern.
This paper investigates the growing role of emotions in shaping policy views. Analyzing social citizens’ media postings and political party messaging over a large variety of policy issues from 2013 to 2024, we document a sharp rise in negative emotions, particularly anger. Content generating anger drives significantly more engagement. We then conduct two nationwide online experiments in the U.S, exposing participants to video treatments that induce positive or negative emotions to measure their causal effects on policy views. The results show that negative emotions increase support for protectionism, restrictive immigration policies, redistribution, and climate policies but do not reinforce populist attitudes. In contrast, positive emotions have little effect on policy preferences but reduce populist inclinations. Finally, distinguishing between fear and anger, we find that anger exerts a much stronger influence on citizens’ policy views, in line with its growing presence in the political rhetoric.
Large language models (LLMs) are increasingly embedded directly into social media platforms, enabling users to request real-time fact-checks of online content. Using an exhaustive dataset of 1,671,841 English-language fact-checking requests made to Grok and Perplexity on X between February and September 2025, we provide the first large-scale empirical analysis of how LLM-based fact-checking operates in the wild. Fact-checking requests comprise 7.6% of all interactions with the LLM bots, and focus primarily on politics, economics, and current events. We document clear partisan asymmetries in usage. Users requesting fact-checks from Grok are much more likely to be Republican than Democratic, while the opposite is true for fact-check requests from Perplexity -- indicating emerging polarization in attitudes toward specific AI models. At the same time, both Democrats and Republicans are more likely to request fact-checks on posts authored by Republicans, and - consistent with prior work using professional fact-checkers and crowd judgments - posts from Republican-leaning accounts are more likely to be rated as inaccurate by both LLMs. Across posts rated by both LLM bots, evaluations from Grok and Perplexity agree 52.6% of the time and strongly disagree (one party rates a claim as true and the other as false) 13.6% of the time. For a sample of 100 fact-checked posts, 54.5% of Grok bot ratings and 57.7% of Perplexity bot ratings agreed with ratings of human fact-checkers, which is significantly lower than the inter-fact-checker agreement rate of 64.0%; but API-access versions of Grok had higher agreement with fact-checkers than did not significantly differ from inter-fact-checker agreement. Finally, in a preregistered survey experiment with 1,592 U.S. participants, exposure to LLM fact-checks meaningfully shifts belief accuracy, with effect sizes comparable to those observed in studies of professional fact-checking. However, responses to Grok fact-checks are polarized by partisanship when model identity is disclosed, whereas responses to Perplexity are not. Together, these findings show that LLM-based fact-checking is rapidly scaling, is generally informative although far from perfect, while also becoming entangled with polarization and partisanship. Our work highlights both the promise and the risks of integrating AI fact-checking into online public discourse.
How does the success of the far right affect the immigration rhetoric of mainstream parties and their electoral performance? To explore this question, we exploit the gradual success of the Front National in French parliamentary elections since its creation in 1972. We find that the electoral success of far-right parties came at the expense of a decline in the vote share of mainstream right-wing parties. Using a comprehensive textual analysis of the universe of candidate manifestos from 1968 to 1997, we show that mainstream right-wing candidates respond to the success of far-right parties by increasing the salience of immigration in their manifestos. This strategic adjustment mitigates electoral losses to far-right competitors. In contrast, the strategic response of mainstream left-wing candidates is to reduce the prominence of immigration in their manifestos to reduce their electoral losses.
Radio music quotas aim to boost exposure to domestic repertoire, but do they shape listening preferences beyond the broadcast context? We investigate whether quota-induced exposure translates into persistent preferences observable on streaming platforms. To do so, we combine an original cross-country database of quota policies with weekly Spotify Top 200 charts spanning 71 countries and individual-level listening histories from Deezer (one of France's leading streaming platforms) matched to survey data on radio exposure. At the aggregate level, quota countries exhibit lower domestic consumption shares on Spotify; however, this pattern likely reflects the endogenous adoption of such policies in markets where domestic content was initially underrepresented. Exploiting the 2024 Greek quota reform as a quasi-experiment, we find no evidence of an increase in national-language streaming consumption, using a synthetic difference-in-differences design. At the individual level, radio listening in France, where a radio quota applies, shows no association with stronger preferences for domestic music once demographic characteristics are accounted for. Overall, our findings provide no evidence that radio music quotas generate meaningful spillover effects on domestic music consumption in the streaming era.
We propose a novel measure of investor attention by analyzing messages sent on Twitter around European Central Bank announcements. We then examine the market impact of the ECB decisions and press conferences, contingent on the level of investor attention prior to the announcements, across a wide array of assets (stocks, bonds, OIS, and exchange rates). We find that absolute price changes are higher when investor attention is elevated before the announcements. This effect is stronger in the press release window than in the press conference window, especially when focusing on messages from users who self-describe themselves as investors. Investor attention also magnifies the impact of monetary policy surprises. Our results suggest that central bankers can use the level of attention prior to the announcements to enhance their anticipation of the market's reaction to the announcement.