This paper studies the generation of inflation expectations using generative AI in survey experiments, specifically focusing on agents created with Large Language Models (LLMs). It reveals that LLMs accessing relevant economic documents exhibit more consistent responses, demonstrating that domain knowledge significantly shapes expectations. Typically, LLMs predict higher inflation than actual rates, aligning closely with findings from the Survey of Consumer Expectations. Information treatments, particularly forward guidance, precisely influence LLMs' inflation expectations. Furthermore, customizing prompts with a persona induces partisan biases, mirroring human survey behaviors. It also explores the importance of model selection and the reasoning behind LLM responses, emphasizing how these elements impact outcomes. Additionally, the paper shows that LLMs can generate granular predictions of inflation expectations across demographic groups, suggesting potential applications in forecasting survey responses with limited data.

This study examines risk and time preferences of large language models (LLMs) in economic decision-making scenarios by adapting established behavioral economics measures. Using a novel methodology combining predefined risk personas and randomized controlled trials (RCTs), we assess AI agents' responses to various economic contexts. We find that AI models demonstrate systematic preference patterns influenced by both assigned personas and economic conditions, with larger models exhibiting more sophisticated preference updating than smaller models. Notably, recession scenarios elicit stronger preference adjustments than boom scenarios across all models, while risk-averse personas maintain consistent positions regardless of context. The results reveal that AI agents can develop coherent economic preferences distinct from human decision-making patterns, particularly in their response to negative economic information.

We simulate economic forecasts of professional forecasters using large language models (LLMs). We construct synthetic forecaster personas using a unique hand-gathered dataset of participant characteristics from the Survey of Professional Forecasters. These personas are then provided with real-time macroeconomic data to generate simulated responses to the SPF survey. Our results show that LLM-generated predictions are similar to human forecasts, but often achieve superior accuracy, particularly at medium- and long-term horizons. We argue that this advantage arises from LLMs' ability to extract latent information encoded in past human forecasts while avoiding systematic biases and noise. Our framework offers a cost-effective, high-frequency alternative that complements traditional survey methods by leveraging both human expertise and AI precision.

We directly compare the persuasion capabilities of a frontier large language model (LLM; Claude Sonnet 3.5) against incentivized human persuaders in an interactive, real-time conversational quiz setting. In this preregistered, large-scale incentivized experiment, participants (quiz takers) completed an online quiz where persuaders (either humans or LLMs) attempted to persuade quiz takers toward correct or incorrect answers. We find that LLM persuaders achieved significantly higher compliance with their directional persuasion attempts than incentivized human persuaders, demonstrating superior persuasive capabilities in both truthful (toward correct answers) and deceptive (toward incorrect answers) contexts. We also find that LLM persuaders significantly increased quiz takers' accuracy, leading to higher earnings, when steering quiz takers toward correct answers, and significantly decreased their accuracy, leading to lower earnings, when steering them toward incorrect answers. Overall, our findings suggest that AI's persuasion capabilities already exceed those of humans that have real-money bonuses tied to performance. Our findings of increasingly capable AI persuaders thus underscore the urgency of emerging alignment and governance frameworks.

Firms' inflation expectations shape pricing, wages, and investment, yet the organizational process through which executive beliefs aggregate into a firm-level forecast remains unmodeled. We develop a multi-agent AI framework in which distinct executive roles produce role-specific inflation expectations, which a Chairman agent synthesizes into a single firm-level expectation. Simulating the Survey of Firms' Inflation Expectations across 32 quarterly waves from 2018 through 2026, automated firms track the main survey dynamics and generate interpretable within-firm disagreement that widens during supply-driven inflation episodes. Role framing changes not only forecast levels but also the economic narratives used to support them. An out-of-sample exercise with staggered model cutoffs suggests that this alignment is not driven solely by memorization. Memory experiments further show that providing the historical path of past SoFIE responses substantially improves accuracy, while feeding back agents' own prior answers introduces anchoring rigidity during regime shifts. In a randomized information-treatment experiment, synthetic firms update beliefs in the direction of the information provided and differentiate between short- and long-horizon signals, demonstrating the framework's potential as a pre-experimental laboratory for information interventions. The framework also reveals systematic limitations: tail-risk probabilities are too high and cross-firm dispersion is substantially compressed. Together, the results show both the promise and the boundaries of multi-agent LLMs as tools for simulating organizational belief formation.

Central banks often sound alike when they discuss inflation and the stance of policy. That similarity could reflect the transmission of language from leading institutions, or a shared response to common global conditions. I score 6,692 monetary policy statements from 51 jurisdictions over 1990-2024 on seven concepts, using a lexical similarity measure that I compare with large language model ratings and text embeddings. Hawkish language and the inflation outlook show the strongest annual comovement across countries, a ranking that survives holding country coverage fixed. This pattern does not carry over consistently to short windows around Federal Reserve releases: the inflation outlook has a small association with what other central banks say afterward under some measures, while the association for hawkish language, the most synchronized annual concept, depends on the scoring method and sample. Widely used measures of international central bank sentiment can therefore reflect common conditions rather than language transmitted from a leading institution.