Chimamanda Ngozi Adichie’s metafictional short story, “Jumping Monkey Hill” is used as a laboratory to explore the relationship between fictionality and believability in a situation where rhetorical cues about a text’s ontological status are misleading and how literary ways of reading might inform engagement with AI texts.
Abstract
A defining if under-acknowledged feature of large language models (LLMs) is their non-indexicality: what AI
engines “know” about the world reflects only their ability to predict and successfully imitate existing textual representations of
that world. Like fictional narratives, therefore, they traffic not in truth or falsity, but believability. Helpfully, literary
critics have experience thinking critically about the ways in which believability is shaped by existing ideologies and biases — an
acknowledged limitation of AI-generated texts as well. Drawing on Neilsen, Phelan and Walsh’s rhetorical model of fictionality,
this article uses Chimamanda Ngozi Adichie’s metafictional short story, “Jumping Monkey Hill” as a laboratory to explore the
relationship between fictionality and believability in a situation where rhetorical cues about a text’s ontological status are
misleading. From there, I suggest how literary ways of reading might inform our engagement with AI texts — including and
especially those that present themselves as factual.
Large language models (LLMs) are rapidly becoming an interface between citizens and political information. They are often regarded as"a better Google."While this analogy might work for some instances, it is unintuitively problematic for democratic politics. A search engine retrieves human-authored documents, while a language model generates novel text that necessarily embeds invisible framing decisions. Because conveying knowledge involves framing, a system that generates answers cannot serve as a neutral conduit to"all human knowledge."Instead, these systems are becoming a new kind of political intermediary. Mechanistic evidence shows that partisan identity is encoded as a locatable geometric direction inside the Llama 3.1 8B model, and that alignment training masks rather than removes this structure. Building on that evidence, we present steering experiments that exploit a model's training cutoff in 2024. This cutpoint auspiciously falls just before a dramatic realignment in American politics marked by the second Trump administration and the MAHA transformation of health politics, providing us with a natural experiment. We find that the model presents temporally contingent partisan alignments as knowledge, with no mechanism for distinguishing fact from opinion. This reality moves the information environment beyond the echo chamber toward an epistemic monoculture where language models, purporting to summarize"all human knowledge"are, in actuality, simply magnifying the cultural and partisan divides inherent in their training data.
A systematic comparison using a naturally occurring online persuasion corpus in which original posters explicitly verify whether a reply changed their view is conducted, highlighting the risk of treating LLM judgments as faithful proxies for human belief updating and point to structural differences in how LLMs and humans process persuasive discourse.
It is argued that LLMs represent a fundamental shift in human-technology interaction, one requiring ritualistic linguistic precision rather than technical commands, and that effective prompt construction functions as ritual magic through three essential elements: clear and specific language, contextual framing, and structured sequencing.
T. Malefyt, Matt Artz· Journal of Business Anthropo...· 1 citation
Can a good argument change an individual’s mind? In three pre-registered experiments, we explore this question in the domain of public sector organizational performance. We observe human subjects as they engage in conversations with a generative artificial intelligence (AI) programmed to argue in one of seven distinct “styles,” including a confrontational challenger style, a didactic style, and a sycophantic style. We develop a theory of effective argumentation predicting that conversational styles which are pleasant and engaging will be more persuasive than styles which are unpleasant or unstimulating. Contrary to this prediction, we find that conversational styles which challenge subjects’ negative views of government agencies produce significant positive attitude change, while sycophantic styles that indulge those views do not. Troublingly, subjects find the sycophantic styles more enjoyable, less frustrating, and more credible than the challenger styles. This dissociation between user experience and persuasive outcome—what we call “grudging persuasion”—suggests that attitude change does not require a pleasant conversational experience, and that the styles subjects enjoy most may be precisely the ones least likely to move them. Our findings point to a potentially dark side of LLM-based persuasion: sycophantic styles that users find most appealing are the least effective at correcting misinformed views.
John D. Marvel, Sheeling Neo, Rachel Cho et al.· Journal of public administra...· 0 citations
It is shown how fact-checking, a generally desirable behavior, can interfere with belief tracking in LLMs and how suppressing this attention at decoding time recovers accuracy only partially and only in some models, calling for future work on intervention methods.
Quang Minh Nguyen, Luis Frentzen Salim· 0 citations
