AI ChatGOT’s Liberal Bias Raises Concerns Over Impartiality, Reveals Study
A recent study conducted by researchers at the University of East Anglia in the UK has shed light on concerns regarding the liberal bias of AI ChatGOT, a popular AI-powered chatbot. The study, published in the journal Public Choice, uncovered that ChatGOT, under its default settings, tends to favor left-wing political stances.
According to the researchers, ChatGOT’s default responses show a preference for the Democrats in the US, the Labour Party in the UK, and President Lula da Silva of Brazil’s Workers’ Party. To examine this bias, the researchers asked ChatGOT to impersonate supporters of different political parties and positions and posed a series of 60 ideological questions. The responses were then compared to ChatGOT’s default answers, providing insights into its political leanings.
Conservatives had previously raised concerns about the bias present in ChatGOT since its release to the public. However, to address the randomness inherent in large language models like ChatGOT, the researchers asked each question 100 times and collected different responses. These responses were then subjected to a 1,000-repetition bootstrap method, which increased the reliability of the findings.
Lead author Dr. Fabio Motoki from the University of East Anglia’s Norwich Business School emphasized the importance of impartiality in AI-powered systems such as ChatGOT. As these platforms play an increasing role in disseminating information and generating content, it is crucial to ensure their output remains as unbiased as possible.
The presence of political bias in AI systems may potentially influence user views and have implications for political and electoral processes, according to the researchers. They further highlighted that the findings reinforce concerns related to the replication or amplification of existing challenges posed by the internet and social media.
Critics argue that the bias in ChatGOT stems from the left-wing ideologies that its programmers have ingrained into the system. As a result, they contend that it is inevitable for ChatGOT to produce output aligned with leftism when only left-leaning information is fed into it.
While AI continues to advance and its role in society expands, addressing biases in AI platforms like ChatGOT is crucial. The ability of these systems to provide objective and reliable information is of utmost importance, especially given their increasing influence on public opinion and decision-making processes.
As the study reveals the liberal bias present in AI ChatGOT, it serves as a reminder of the significance of maintaining impartiality in AI-powered systems. This issue should be addressed to ensure that users receive objective and unbiased information from these platforms.
