E85: Discourse and Manipulation

https://is1-ssl.mzstatic.com/image/thumb/Podcasts125/v4/93/d8/e5/93d8e5c4-a23a-8cdf-f87d-7dc3a73cee11/mza_674862149011135675.png/600x600bb.jpg

re:verb

Calvin Pollak and Alex Helberg

100 episodes

1 month ago

On today's show, Alex and Calvin continue their series on “AI” and public discourse, focusing this time on the increasing proliferation of AI applications in government writing, policy, and social media. We characterize the second Trump administration as the "first totally post-AI presidency," which has adopted the "dumbest, most unreflective, most uncritical approach" to AI's use in communication, research, and analysis. Throughout the show, we emphasize how AI technologies are themselves rhetorical artifacts at the same time as they so often produce “bad” rhetoric, reflecting the intentions, values, and presuppositions of their creators, as well as the inherent biases of their training data and text generation models. This often results in an entry-level, overly dense writing style - often referred to as "slop" - which is almost written not to be read, but rather to fill space. We explore several concerning examples of AI's uncritical adoption by the secondTrump administration and their acolytes in the tech world. Early executive orders exhibited AI-generated formatting errors and formulaic, generic language, demonstrating a context-blind style that could lead to legal problems and erode public trust. Furthermore, the "MAHA Report" from the Office of Health and Human Services was found to fabricate studies and misrepresent findings, reflecting how large language models are "sycophantic," and can reinforce existing (often false) beliefs. Our discussion also covers Palantir's "Foundry" product, which aims to combine diverse government datasets, raising significant privacy and political concerns, especially given the political leanings of Palantir’s founders. Finally, we examine xAI’s Grok chatbot (run by Elon Musk), which illustrates how tech elites can exert incredible political power through direct interventions in AI tools’ system prompts - which in recent months has led Grok to parrot conspiracy theories and make explicit antisemitic remarks on the public feeds of X/Twitter. Ultimately, our analyses emphasizes - once again - that these so-called “AI” technologies are not neutral; they are, in the words of Matteo Pasquinelli, "crystallization[s] of a productive social process" that "reinforce the power structure that underlies [them]," perpetuating existing inequalities. Understanding these mechanisms and engaging in what Pasquinelli terms "de-connectionism" - undoing the social and economic fabric constituting these systems - is essential for critiquing the structural factors and power dynamics that AI reproduces in public discourse. Have any questions or concerns about this episode? Reach out to our new custom-tuned chatbot, @Bakh_reverb on X/Twitter! Examples Analyzed in this Episode: Trump Admin Accused of Using AI to Draft Executive Orders https://www.yahoo.com/news/trump-admin-accused-using-ai-191117579.html Eryk Salvaggio - “Musk, AI, and the Weaponization of ‘Administrative Error’” https://www.techpolicy.press/musk-ai-and-the-weaponization-of-administrative-error/ Emily Kennard & Margaret Manto (NOTUS) - “The MAHA Report Cites Studies That Don’t Exist” - https://archive.ph/WVIrT Sheera Frenkel & Aaron Krolik (NYT) - “Trump Taps Palantir to Compile Data on Americans” https://www.nytimes.com/2025/05/30/technology/trump-palantir-data-americans.html David Klepper - “Gabbard says AI is speeding up intel work, including the release of the JFK assassination files” https://apnews.com/article/gabbard-trump-ai-amazon-intelligence-beca4c4e25581e52de5343244e995e78 Miles Klee - “Elon Musk’s Grok Chatbot Goes Full Nazi, Calls Itself ‘MechaHitler’” - https://archive.ph/SdoJn Works & Concepts Cited in this Episode: Bakhtin, M. M. (2010). The dialogic imagination: Four essays. University of Texas Press. Benjamin, R. (2019). Race after technology: Abolitionist tools for the new Jim code (1st ed.). Polity. Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021, March). On the dangers of stochastic parrots: Can language models be too big?. In Proceedings of the 2021 ACM conference on fairness, accountability, and transparency (pp. 610-623). Our previous episode with Dr. Bender about her work Burke, K. (1984). Permanence and change: An anatomy of purpose. Univ of California Press. Burke, K. (1965). Terministic screens. In Proceedings of the American Catholic philosophical association (Vol. 39, pp. 87-102). DeLuca, L. S., Reinhart, A., Weinberg, G., Laudenbach, M., Miller, S., & Brown, D. W. (2025). Developing Students’ Statistical Expertise Through Writing in the Age of AI. Journal of Statistics and Data Science Education, 1-13. Haggerty, K. D., & Ericson, R. V. (2017). The surveillant assemblage. Surveillance, crime and social control, 61-78. Hill, K. (2025, 13 June). “They Asked an A.I. Chatbot Questions. The Answers Sent Them Spiraling.” The New York Times. Markey, B., Brown, D. W., Laudenbach, M., & Kohler, A. (2024). Dense and disconnected: Analyzing the sedimented style of ChatGPT-generated text at scale. Written Communication, 41(4), 571-600. Miller, C. R. (1984). Genre as social action. Quarterly journal of speech, 70(2), 151-167. Murakami, H. (1994). Dance dance dance : a novel (1st ed.). Kodansha International. Pasquinelli, M. (2023). The eye of the master: A social history of artificial intelligence. Verso Books. Reinhart, A., Markey, B., Laudenbach, M., Pantusen, K., Yurko, R., Weinberg, G., & Brown, D. W. (2025). Do LLMs write like humans? Variation in grammatical and rhetorical styles. Proceedings of the National Academy of Sciences, 122(8), e2422455122. An accessible transcript for this episode can be found here (via Descript)

All content for re:verb is the property of Calvin Pollak and Alex Helberg and is served directly from their servers with no modification, redirects, or rehosting. The podcast is not affiliated with or endorsed by Podjoint in any way.

E85: Discourse and Manipulation

re:verb

1 hour 22 minutes 35 seconds

1 year ago

E85: Discourse and Manipulation

On today’s show, Alex and Calvin break down the concept of “Manipulation” in rhetoric and political discourse. We outline some key strategies for identifying and critiquing manipulation, and discuss its social and political implications as a form of large-scale “mind control.” The term manipulation, as we define it, comes from a school of linguistic and discourse analysis known as Critical Discourse Studies (CDS), which is primarily concerned with the ways language is used to reinforce inequality and power differentials in society. We walk through how the term is defined by CDS scholar Teun van Dijk, from his landmark 2006 article “Discourse and Manipulation.” In it, van Dijk gives us a toolkit for understanding 3 different levels of manipulation: (1) social, which designates the human relationships, power positions, and organizational and political resources required to effect manipulation at scale; (2) cognitive, which designates how manipulative language forms mental models that influence people’s thoughts and actions in the world; and (3) discursive, which captures the various linguistic, stylistic, and rhetorical strategies that tend to recur in manipulation. To put this term in context, we analyze an example of discourse manipulation surrounding student protests against the most recent flare-up in Israel’s war on Gaza: Carnegie Mellon University President Farnam Jahanian’s email to university students and faculty in response to a recent student vigil where the phrase “from the river to the sea” was chanted. We closely analyze the careful manipulations of emphasis and value that Jahanian creates in his discourse, which subtly demonizes student demonstrators advocating for peace and the cessation of violence between Israel and Hamas, while reaffirming the supposedly apolitical “commitments” of the institution he represents. Full Text Version of Farnam Jahanian Email Works and Concepts Cited in this Episode: Fairclough, N. (2003). Analysing discourse: Textual analysis for social research. Psychology Press. McGee, M. C. (1980). The “ideograph”: A link between rhetoric and ideology. Quarterly journal of speech, 66(1), 1-16. [Our 2018 re:blurb on Ideographs can be found here.] Oddo, J. (2019). The discourse of propaganda: Case studies from the Persian Gulf War and the War on Terror. Penn State University Press. [Our September 2021 episode with CDS scholar John Oddo can be found here.] van Dijk, T. A. (2006). Discourse and manipulation. Discourse & society, 17(3), 359-383. An accessible transcript of this episode is available upon request. Please reach out to us via email (reverbcontent@gmail.com), social media, or our website contact form to request a transcript.

re:verb