{"id":325556,"date":"2025-12-04T20:37:08","date_gmt":"2025-12-04T20:37:08","guid":{"rendered":"https:\/\/www.newsbeep.com\/ca\/325556\/"},"modified":"2025-12-04T20:37:08","modified_gmt":"2025-12-04T20:37:08","slug":"chatbots-can-sway-political-opinions-but-are-substantially-inaccurate-study-finds-artificial-intelligence-ai","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/ca\/325556\/","title":{"rendered":"Chatbots can sway political opinions but are \u2018substantially\u2019 inaccurate, study finds | Artificial intelligence (AI)"},"content":{"rendered":"

Chatbots can sway people\u2019s political opinions but the most persuasive artificial intelligence models<\/a> deliver \u201csubstantial\u201d amounts of inaccurate information in the process, according to the UK government\u2019s AI security body.<\/p>\n

Researchers said the study was the largest and most systematic investigation of AI persuasiveness to date, involving nearly 80,000 British participants holding conversations with 19 different AI models.<\/p>\n

The AI Security Institute<\/a> carried out the study amid fears that chatbots can be deployed for illegal activities including fraud and grooming.<\/p>\n

The topics included \u201cpublic sector pay and strikes\u201d and \u201ccost of living crisis and inflation\u201d, with participants interacting with a model \u2013 the underlying technology behind AI tools such as chatbots \u2013 that had been prompted to persuade the users to take a certain stance on an issue.<\/p>\n

Advanced models behind ChatGPT and Elon Musk\u2019s Grok were among those used in the study, which was also authored by academics at the London School of Economics, Massachusetts Institute of Technology, the University of Oxford and Stanford University.<\/p>\n

Before and after the chat, users reported whether they agreed with a series of statements expressing a particular political opinion.<\/p>\n

The study<\/a>, published in the journal Science on Thursday, found that \u201cinformation-dense\u201d AI responses were the most persuasive. Instructing the model to focus on using facts and evidence yielded the largest persuasion gains, the study said. However, the models that used the most facts and evidence tended to be less accurate than others.<\/p>\n

\u201cThese results suggest that optimising persuasiveness may come at some cost to truthfulness, a dynamic that could have malign consequences for public discourse and the information ecosystem,\u201d said the study.<\/p>\n

On average, the AI and human participant would exchange about seven messages each in an exchange lasting 10 minutes.<\/p>\n

It added that tweaking a model after its initial phase of development, in a practice known as post-training, was an important factor in making it more persuasive. The study made the models, which included freely available \u201copen source\u201d models such as Meta\u2019s Llama 3 and Qwen by the Chinese company Alibaba, more convincing by combining them with \u201creward models\u201d that recommended the most persuasive outputs.<\/p>\n

Researchers added that an AI system\u2019s ability to churn out information could make it more manipulative than the most compelling human.<\/p>\n

\u201cInsofar as information density is a key driver of persuasive success, this implies that AI could exceed the persuasiveness of even elite human persuaders, given their unique ability to generate large quantities of information almost instantaneously during conversation,\u201d said the report.<\/p>\n

Feeding models personal information about the users they were interacting with did not have as big an impact as post-training or increasing information density, said the study.<\/p>\n

Kobi Hackenburg, an AISI research scientist and one of the report\u2019s authors, said: \u201cWhat we find is that prompting the models to just use more information was more effective than all of these psychologically more sophisticated persuasion techniques.\u201d<\/p>\n

However, the study added that there were some obvious barriers to AIs manipulating people\u2019s opinions, such as the amount of time a user may have to engage in a long conversation with a chatbot about politics. There are also theories suggesting there are hard psychological limits to human persuadability, researchers said.<\/p>\n

Hackenburg said it was important to consider whether a chatbot could have the same persuasive impact in the real world where there were \u201clots of competing demands for people\u2019s attention and people aren\u2019t maybe as incentivised to sit and engage in a 10-minute conversation with a chatbot or an AI system\u201d.<\/p>\n","protected":false},"excerpt":{"rendered":"Chatbots can sway people\u2019s political opinions but the most persuasive artificial intelligence models deliver \u201csubstantial\u201d amounts of inaccurate…\n","protected":false},"author":2,"featured_media":325557,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[62,276,277,49,48,61],"class_list":["post-325556","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-ai","tag-artificial-intelligence","tag-artificialintelligence","tag-ca","tag-canada","tag-technology"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/posts\/325556","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/comments?post=325556"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/posts\/325556\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/media\/325557"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/media?parent=325556"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/categories?post=325556"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/tags?post=325556"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}