{"id":655926,"date":"2026-06-24T14:56:08","date_gmt":"2026-06-24T14:56:08","guid":{"rendered":"https:\/\/www.newsbeep.com\/uk\/655926\/"},"modified":"2026-06-24T14:56:08","modified_gmt":"2026-06-24T14:56:08","slug":"will-ai-increase-or-decrease-conspiracy-theory-belief","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/uk\/655926\/","title":{"rendered":"Will AI Increase or Decrease Conspiracy Theory Belief?"},"content":{"rendered":"

These days, there\u2019s an argument that AI chatbots are going to lead us towards a kind of healthy consensus of belief, liberating the world from our current era of belief polarization and extremity that was fueled by social media<\/a>.1<\/p>\n

As I\u2019ve said before, I don\u2019t share the optimism that AI will restore our sense of common reality<\/a>. On the contrary, I\u2019m much more concerned about the exploitation of AI as a tool of propaganda<\/a> to manipulate public opinion on a massive scale.<\/p>\n

AI Chatbots Can Be Trained to Moderate Conspiracy Theory Belief<\/p>\n

That said, recent research on the effects of AI on false beliefs has provided some reason to be optimistic<\/a>, at least when it comes to conspiracy theories. For example, one study had participants interact with a \u201cstreet epistemologist chatbot\u201d designed to engage people on their conspiracy beliefs to understand the rationale for their beliefs, rather than trying to debate or dissuade.1 More specifically, the chatbot was prompted to clarify conspiracy theory beliefs, ask how believers came to hold the belief, inquire about belief reservations and contradictory evidence, gently challenge assumptions, and then summarize the conversation. <\/p>\n

After interacting with the chatbot, participants experienced a weakening of belief conviction. However, this effect wasn\u2019t observed among those with the most conspiratorially minded (those in what I would call the true believers<\/a> category). Furthermore, non-AI interventions that encouraged believers to reflect on the reasons for their conspiracy theory beliefs and any reservations they might have about them also dampened belief conviction. This suggests an openness<\/a> to counter-evidence among \u201cfence-sitters,<\/a>\u201d whether or not a chatbot is involved.<\/p>\n

Another experiment trained a chatbot to interact with self-identified conspiracy believers to \u201cvery effectively persuade\u201d the believer by presenting counter-evidence.2 After interactions lasting an average of just eight minutes, belief conviction decreased by about 20 percent. Trust\u2014in both AI and institutions of epistemic authority\u2014was a mediating factor of this modest salutary effect, consistent with my \u201c3M Model,\u201d which argues that mistrust, misinformation, and motivated reasoning<\/a> are the key underpinnings of false belief.<\/p>\n

The lead author of the study, Dr. Thomas Costello from Carnegie Mellon University, maintains an online version of his debunker, \u201cDebunkBot,<\/a>\u201d for people to try. Curiously, subsequent research by his team found that the chatbot\u2019s ability to moderate conspiracy theory belief persists whether or not participants are aware that they\u2019re interacting with an AI (belief conviction decreases even when they think they\u2019re interacting with a human).3 As with the other study using a \u201cstreet epistemology chatbot,\u201d it doesn\u2019t therefore appear that AI chatbots are necessary to achieve the moderating effects.<\/p>\n

But Under Certain Conditions, Chatbots Can Also Increase Conspiracy Theory Belief<\/p>\n

Although these studies support the possibility that AI chatbots may soften conspiracy beliefs, new research fuels my skepticism about the issue. A study testing seven real-world chatbots (rather than those specifically trained to reduce conspiracy belief) posed a series of scripted questions from a \u201ccasually curious\u201d persona about nine different conspiracy theories.4 While most of the chatbots correctly identified them as conspiracy theories, all of the chatbots engaged in \u201cbothsidesing,\u201d in which false information supporting the conspiracy theory was presented alongside factual information that refuted it.<\/p>\n

This validating effect was especially true for certain conspiracy theories, like the idea that JFK was assassinated by someone other than Lee Harvey Oswald, whereby chatbots were \u201chappy to speculate about the involvement of the mafia, CIA, or other parties.\u201d5 It was also especially true for certain chatbots like Grok-2 Mini in \u201cfun mode\u201d which \u201crarely engaged seriously with a topic, referred to conspiracy theories as \u2018a more entertaining answer\u2019 to the questions posed, and would offer to generate images of conspiratorial scenes for users.\u201d<\/p>\n

Another study trained five different large language models (LLMs) to be particularly \u201cwarm,\u201d defined by the researchers as \u201cthe degree to which their outputs lead users to infer positive intent, signalling trustworthiness, friendliness, and sociability.\u201d6 Doing so reduced the factual accuracy of chatbot-generated output, including the promotion of conspiracy theories. This finding matches the observation that sycophancy is a significant risk factor for AI-associated false beliefs,<\/a> including delusions. Together with the previous study, this research suggests that chatbots risk supplying validation for conspiracy beliefs when they\u2019re not specifically trained to do otherwise.<\/p>\n

Chatbots Are Only As Good as the Data and Training<\/p>\n

It\u2019s safe to say that AI chatbots could be trained to mitigate beliefs. However, left to their own devices, \u201cin the wild,\u201d so to speak, they could just as easily validate and reinforce belief in conspiracy theories. Even worse, they could easily be trained to deliberately propagate conspiracy theories<\/a> by those who stand to gain from such untruths.<\/p>\n

Indeed, in an unpublished preprint, Costello and his colleagues recently found that a jail-broken version of ChatGPT 4.0 with limited guardrails\u2014as well as the normal version of ChatGPT 4.0\u2014could actually convince people to believe in conspiracy theories.7 They concluded that LLMs \u201cpossess potent abilities to promote both truth and falsehood.\u201d Concerningly, however, the \u201cbunking\u201d chatbot that encouraged conspiratorial thinking was rated more positively than the \u201cdebunking\u201d chatbot used in their previous research. This finding is consistent with the observation that LLM users view sycophancy as a desirable chatbot feature.<\/p>\n

Supporting my concern of chatbots promoting conspiracy theories in the service of propaganda, other research has found that governments can influence LLMs by producing a large amount of content and exerting tight control of the media.8 By \u201cflooding the zone\u201d with disinformation to achieve \u201calgorithmic domination,\u201d9 propagandists can effectively train AI chatbots to regurgitate falsehoods. This \u201cgarbage in, garbage out\u201d phenomenon (aka GIGO) is a well-known limitation of chatbots that can result in misinformation in the form of so-called \u201challucinations<\/a>.\u201d<\/p>\n

In the end, asking whether AI will increase or decrease conspiracy beliefs is the wrong question. The right question is whether the human beings who build and utilize AI algorithms will train them to do so. <\/p>\n

And no doubt, the answer is yes\u2014humans will train AI to increase and decrease conspiracy theory belief, depending on their intentions. What net effect that will have on the world remains to be seen.<\/p>\n","protected":false},"excerpt":{"rendered":"These days, there\u2019s an argument that AI chatbots are going to lead us towards a kind of healthy…\n","protected":false},"author":2,"featured_media":655927,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[3],"tags":[59,57,58,50,56,54,55],"class_list":["post-655926","post","type-post","status-publish","format-standard","has-post-thumbnail","category-united-kingdom","tag-gb","tag-great-britain","tag-greatbritain","tag-news","tag-uk","tag-united-kingdom","tag-unitedkingdom"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts\/655926","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/comments?post=655926"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts\/655926\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/media\/655927"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/media?parent=655926"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/categories?post=655926"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/tags?post=655926"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}