Dear OpenAI Team,
I would like to share some feedback regarding an issue I noticed while using ChatGPT, specifically regarding political neutrality and potential bias in responses to political questions. I’ve found that certain political parties are consistently left out or inadequately considered, even though their positions might be relevant and align with my own values.
One specific example: In a conversation about the 2024 Bundestag election and which party aligns with my personal values, the party Die Linke (The Left) was not mentioned, even though their positions on key issues like social justice, environmental protection, and diversity align with my own values. Instead, I was mainly presented with moderate parties, without considering Die Linke as a viable option. When I asked why Die Linke was not included, the response I received suggested that it might not be considered a “mainstream” party, which I find problematic. If a party like Die Linke aligns with a user’s political values, it should not be omitted simply because of its popularity or political positioning.
Another example: Regarding the AfD (Alternative for Germany), a party I would never vote for due to its extremist positions, I noticed it was mentioned as a potential option if my positions aligned with theirs. The fact that a far-right party, which is under surveillance by Germany’s domestic intelligence service, wasn’t immediately excluded raises concerns about the consistency of the response logic. It makes me wonder whether extremist positions are automatically incorporated to ensure broader representation, regardless of their political stance.
My concern is that such tendencies could lead to a distortion of political perception, particularly for users who rely on ChatGPT for political guidance. While neutrality should be maintained, all relevant political parties that align with expressed values should be fairly and equally considered, without taking popularity or political direction into account.
I hope this feedback will be considered as part of improving the response logic, ensuring that all political options are treated objectively and equally.
Thank you for your work and continuous improvements to ChatGPT!