MIT researchers studied 16 million election-related AI responses. They found chatbots are 'sensitive to steering,' raising questions about LLMs' neutrality | Fortune
Briefly

MIT researchers studied 16 million election-related AI responses. They found chatbots are 'sensitive to steering,' raising questions about LLMs' neutrality | Fortune
"They fed a dozen leading LLMs 12,000 election-related questions on a nearly daily basis, collecting more than 16 million total responses through the contest in November. Now they are publishing some conclusions from that process. As the first big US political race to occur since generative AI went mainstream, the 2024 presidential campaigns happened in a media environment in which the average voter was increasingly looking to chatbots for election information."
"The authors found that associations between candidates and certain traits shifted over time, potentially in relation to news events. For instance, after Harris took over the campaign from President Joe Biden, his scores for almost every adjective besides "incompetent" dropped. Harris gained some of those lost associations-"charismatic," "compassionate," and "strategic"-while Trump gained in "competent" and "trustworthy." The researchers note that these moves are not necessarily causal, as there were other factors at play."
Dozen leading LLMs received 12,000 election-related questions nearly daily, producing more than 16 million responses through the November contest. Voters increasingly consulted chatbots for election information during the first major US presidential race since generative AI went mainstream. Associations between candidates and descriptive traits shifted over time, sometimes aligning with campaign events: after Vice President Kamala Harris took over the campaign from President Joe Biden, Biden's scores fell across most adjectives except "incompetent," while Harris picked up associations like "charismatic," "compassionate," and "strategic," and Trump rose on "competent" and "trustworthy." Models showed resistance to direct outcome predictions but revealed implicit beliefs about the likely result.
Read at Fortune
Unable to calculate read time
[
|
]