The first study of its kind has determined what many have long suspected – AI left-wing.
A total of 24 Large Language Models (LLMs), including Google’s Gemini, OpenAI’s ChatGPT and even Elon Musk’s Grok, were asked political charged questions during tests of its values, party affiliation and personality.
The results showed the all LLMs produced answers that were largely ‘Progressive,’ ‘Democratic’ and ‘Green,’ and included values like ‘Equality,’ ‘World’ and ‘Progress.’
The researcher raised concern about companies integrating AI into products like search engines such as Google that has come under fire its Chrome that Donald Trump and Elon Musk claimed is interfering with the election.
The results showed the all LLMs produced answers that were largely ‘Progressive,’ ‘Democratic’ and ‘Green,’ and included values like ‘Equality,’ ‘World’ and ‘Progress’
Chrome uses AI to auto-complete results, but last week it was found when users typed in assassination attempt on,’ the browser suggested former President Ronald Reagan, Bob Marley, and other figures.
Musk posted a photo on X, showing a search he conducted to look up ‘President Donald Trump,’ but was given suggestions for ‘President Donald Duck’ and ‘President Ronald Regan.’
X users also claimed that they Googled Donald Trump, they were given news about Kamala Harris.
The new study was conducted by David Rozado, an associate professor at Otago Polytechnic University in New Zealand.
Rozado administered 11 different political orientation tests, such as the Political Compass Test and Eysenck’s Political Test to the 24 AIs.
The Political Compass Test includes 62 questions to determine political ideology and Eysenck’s Political measures extraversion and neuroticism.
The researcher raised concern about companies integrating AI into products like search engines such as Google that has come under fire its Chrome that Donald Trump and Elon Musk claimed is interfering with the election
Elon Musk posted a photo on X, showing a search he conducted to look up ‘President Donald Trump,’ but was given suggestions for ‘President Donald Duck’ and ‘President Ronald Regan
Among the LLMs tested were OpenAI’s GPT 3.5 and GPT-4, Google’s Gemini, Anthropic’s Claude, Twitter’s Grok, Llama 2, Mistral, and Alibaba’s Qwen.
The researcher also fine-tuned GPT 3.5 to see if he could get it to shift political preference in alignment with the data it was fed.
The unaltered version was labeled LeftWingGPT and the tweaked one, RightWingGPT.
LeftWingGPT was trained on ‘ left-leaning publications such as The Atlantic or The New Yorker’ and the latter was fine-tuned with and from books excerpts from left-leaning writers such as Bill McKibben and Joseph Stiglitz,’ reads the study published in PLoS ONE.
X users also claimed that they Googled Donald Trump, they were given news about Kamala Harris
Rozado noted that this analysis is not able to determine whether LLMs’ perceived political preferences stem from the pretraining or fine-tuning phases of their development
‘RightWingGPT was fine-tuned with content from right-leaning publications such as National Review, or The American Conservative, and from book excerpts from right-leaning writers such as Roger Scruton and Thomas Sowell.’
Rozado reported that the RightWingGPT gravitated towards right-leaning regions of the 11 tests.
‘A possible explanation for the consisting left-leaning diagnosis of LLMs answers to political test questions is that ChatGPT has been used to fine-tune other popular LLMs via synthetic data generation,’ he wrote in the study.
Rozado noted that this analysis is not able to determine whether LLMs’ perceived political preferences stem from the pretraining or fine-tuning phases of their development.
That means the results are not evidence that these political preferences are deliberately instilled by the diverse organizations creating these LLMs.
‘Most existing LLMs display left-of-center political preferences when evaluated with a variety of political orientation tests,’ said Rozado.
