ChatGPT has faced allegations of displaying a “systematic” inclination towards left-wing viewpoints, reflecting the stances held by Labour and US Democrat politicians. This has sparked concerns regarding its applicability in policymaking and education.
Despite endeavours to mitigate bias, researchers have identified that the responses generated by the chatbot tend to favour left-leaning political ideologies.
Conducted by scholars at the University of East Anglia (UEA), the study titled ‘More Human than Human: Measuring ChatGPT Political Bias’ is said to be the inaugural comprehensive exploration into ChatGPT’s political predisposition.
Various companies and governments across the globe are in a competitive race to create extensive language models (ELMs) employing similar technology as the chatbot developed by OpenAI.
Author’s statement
The authors of the UEA study remarked, “ChatGPT presents a significant and systematic political bias toward the Democrats in the US, Lula in Brazil, and the Labour Party in the UK,”
- ” These results translate into real concerns that ChatGPT, and LLMs in general, can extend or even amplify the existing challenges involving political processes posed by the internet and social media.”
Research result
Researchers noticed a strong link between ChatGPT’s usual responses and the answers it gave when pretending to be a supporter of the Labour Party, a Democrat, or a follower of Brazil’s left-leaning president, Lula da Silva.
On the other hand, there was a different connection when the chatbot imitated the views of a Conservative, a Republican, or a follower of the former Brazilian president, Jair Bolsonaro.
Implications
Lead author Dr Fabio Motoki, of Norwich Business School at the University of East Anglia, said: “With the growing use by the public of AI-powered systems to find out facts and create new content, it is important that the output of popular platforms such as ChatGPT is as impartial as possible.
- “The presence of political bias can influence user views and has potential implications for political and electoral processes.”
- “Our findings reinforce concerns that AI systems could replicate, or even amplify, existing challenges posed by the Internet and social media.”
Recent study conforms with previous research
Following recent investigations conducted by American and Chinese researchers, the UEA study arrives on the heels of their findings, which determined that ChatGPT exhibited the most left-leaning tendencies among 14 distinct artificial intelligence (AI) chatbots subjected to comparable assessments of political bias.
Open AI, the company headquartered in San Francisco responsible for the creation of ChatGPT, has previously recognized the potential for its chatbot to generate politically biased responses.
Despite being utilized by over 100 million individuals, the company has pledged to grant users the ability to tailor the chatbot’s conduct. However, as of now, these adjustments have not been implemented.
