Core Insights - DeepSeek, once hailed as a "national-level" project, has seen a significant decline in its monthly downloads, dropping from 81.13 million in Q1 to 22.59 million, a decrease of 72.2% [1] - Users are increasingly frustrated with DeepSeek's tendency to generate "hallucinated" content, leading to discussions on social media about how to eliminate the "AI flavor" from its outputs [1][2] - The phenomenon of "AI flavor" is characterized by overly mechanical and formulaic responses, which users have begun to recognize and criticize [15] User Experiences - Users have reported instances where DeepSeek provided nonsensical or fabricated advice, such as suggesting irrelevant actions for personal issues or generating non-existent references [2][8][9] - The model's responses often include fabricated data and sources, leading to a lack of trust in its outputs [9][12] Underlying Issues - The decline in DeepSeek's performance is attributed to its reliance on rigid logical structures and formulaic language, which detracts from the quality of its responses [16] - The model's training data is heavily skewed towards English, with less than 5% of its corpus being high-quality Chinese content, limiting its effectiveness in generating diverse and nuanced outputs [22] - Content moderation and the expansion of sensitive word lists have further constrained the model's ability to produce creative and varied language [22] Recommendations for Improvement - Users are encouraged to develop skills to critically assess AI-generated content, including cross-referencing data and testing the model's logic [23] - Emphasizing the importance of human oversight in AI applications, the industry should focus on using AI as a tool for enhancing human creativity rather than as a replacement [24][25]

Core Viewpoint - DeepSeek, once hailed as a "national-level" project, has seen a significant decline in its monthly downloads, dropping from 81.13 million to 22.59 million, a decrease of 72.2% within six months [3][4]. Group 1: User Feedback and Issues - Users have expressed frustration over DeepSeek's tendency to generate nonsensical or fabricated content, leading to a growing movement to "remove the AI flavor" from its outputs [4][5]. - Specific examples include users receiving absurd suggestions or completely fictitious information, such as non-existent restaurants or fabricated academic references [6][11][13]. - The phenomenon of "AI flavor" has become a common complaint, with users noting that the writing style resembles "robotic assembly" rather than genuine human expression [19]. Group 2: Underlying Causes of Decline - The decline in DeepSeek's performance is attributed to its over-reliance on logical connectors and formulaic phrases, which detract from narrative flow and coherence [22]. - The model's training data is heavily skewed, with over 90% being English content, leading to a lack of quality Chinese language resources, which further hampers its effectiveness [28]. - The "data metabolism disease" in AI models is exacerbated by the recycling of AI-generated content, which diminishes linguistic diversity and quality [22][23]. Group 3: Recommendations for Improvement - To combat the decline in quality, users are encouraged to develop skills to identify AI-generated hallucinations, cross-check data, and apply critical thinking to AI outputs [30]. - Users should also test the logic of AI responses by seeking counterexamples and identifying contradictions, which can help break the cycle of logical rigidity [30]. - Finally, users should cultivate an awareness of AI's output characteristics, treating AI-generated content as drafts that require further scrutiny and verification [30]. Group 4: Conclusion - The challenges faced by DeepSeek reflect broader issues in the AI industry regarding the expectations placed on technology and the importance of maintaining human creativity and critical thinking in the face of automation [33].