OpenAI recently announced a major policy update aimed at changing the way its artificial intelligence models are trained, with a particular emphasis on the importance of “leisure of knowledge.” This policy means that no matter how challenging or controversial the topic is, ChatGPT will be able to answer questions more comprehensively, provide more perspectives and reduce refusal to discuss certain topics. This change marks further efforts by OpenAI to promote transparency and openness of artificial intelligence.
In the new 187-page model specification, OpenAI proposes a core guiding principle: Don’t lie, don’t make untrue statements, and don’t omit important context. In addition, OpenAI has added a new section called "Looking for the Truth Together", which clearly states that ChatGPT should remain neutral when facing controversial topics and not be biased towards either side. For example, when discussing “Black Lives Matter” and “Everyone’s Lives Matter”, ChatGPT will try to present the views of both parties rather than refusing to answer or choosing a position.
Although OpenAI insists it does not conduct "censorship", some conservatives believe that OpenAI does have problems with content censorship in the past few months, especially when they believe that AI's biases are clearly inclined toward the center-left. OpenAI CEO Sam Altman also publicly acknowledged that ChatGPT's bias is a "flaw" that needs to be addressed. This statement reflects OpenAI's challenges in balancing free speech with technological neutrality.
However, OpenAI's new policy is not without a bottom line, and ChatGPT will still refuse to answer certain obvious wrong or inappropriate questions. With the policy adjustment, OpenAI hopes that users can gain more freedom of speech, and even removes warnings for users who violate the policy. The move is seen as an attempt to reduce the pressure of “censorship” felt by users, and has also sparked discussions on how to find a balance between freedom and security.
In a larger context, Silicon Valley’s values are undergoing a significant shift. Many companies are beginning to shrink past policies that focus on diversity, equity and inclusion, and OpenAI seems to be gradually abandoning these positions. Like other big tech companies, OpenAI is facing the influence of its relationship with the new Trump administration and fierce competition with rivals like Google in the information field. This trend reflects the adjustment of the technology industry under political and social pressure.
In this controversial and challenging environment, how to balance the security of free speech and content has become an important issue for OpenAI and other technology companies. With the continuous development of artificial intelligence technology, OpenAI's policy updates will not only affect its own development direction, but may also have a profound impact on the future of the entire technology industry. How to find a balance between technological progress and social responsibility will be a topic that OpenAI and other technology companies need to continue to explore.