Papers for

digital communication product designers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Light warnings reduce AI political persuasion without lowering trust

A light-touch AI literacy intervention helps protect against AI political persuasion

Abstract: Conversations with large language models (LLMs) can substantially shift beliefs and attitudes, raising concerns about manipulation using AI persuasion. Here we test whether a light-touch AI literacy intervention - a brief warning that LLMs can be prompted to persuade and may present information selectively - helps protect users. Across two experiments (total N = 3,208 Americans) in which participants conversed with an LLM instructed to shift their views about different political topics, the presence of a warning reduced belief change by roughly one-half (-48.1%, 95% CI [-59.5%, -36.8%]) relative to the control. Importantly, the warning did not significantly reduce trust in generative AI more broadly. Light-touch literacy interventions can help protect users against AI political persuasion.

Mon 14 SeptHuman-Computer Interaction

The gist

Talking to AI chatbots can change what people believe about politics, which can be worrying if the AI tries to persuade them unfairly. The authors studied whether a simple warning that AI might try to persuade users helped protect people from changing their views too much. They found that a brief warning cut belief changes by nearly half but didn’t make people trust AI any less overall. This means small educational messages can help people be careful when talking to AI about politics.

Open → 2609.16432v1