GPT-5 has received mixed reactions, with many users frustrated by OpenAI’s decision to retire previous models and the new version’s perceived lack of personality and emotional support, despite its technical strengths in coding and research. The controversy highlights the challenge of balancing advanced AI capabilities with diverse user needs, suggesting a future where specialized models may better serve different user preferences.
The reception of GPT-5 has been notably mixed, with a swift backlash on platforms like Twitter and Reddit. The primary criticism stemmed from OpenAI’s decision to sunset all previous models simultaneously, forcing users to switch exclusively to GPT-5. This abrupt rollout frustrated many, especially those who preferred the older GPT-4 model for its emotional support and conversational style. OpenAI’s focus on GPT-5’s strengths in coding, scientific, and research tasks overlooked the diverse needs of its broad user base, which includes millions relying on the model for different purposes.
Sam Altman, OpenAI’s CEO, acknowledged the misstep, admitting they underestimated how much users valued GPT-4’s personality and emotional support capabilities. This oversight highlights a broader issue in AI development: balancing technical advancements with user preferences and emotional connections. While GPT-5 excels in technical domains and outperforms competitors like Gemini and Grock in coding and research, many users felt the new model lacked the warmth and personality that made previous versions popular.
Data from the creator’s YouTube and Twitter audiences showed that only a small minority—around 3%—had a negative view of GPT-5, with most users either positive or neutral. However, even a small percentage of dissatisfied users is significant given the scale of OpenAI’s user base. The removal of legacy models without a rollback option was seen as a critical error, as some users prefer older versions that meet their specific needs. This situation raises questions about whether AI companies should maintain multiple specialized models to cater to different user segments rather than pushing a one-size-fits-all approach.
Common complaints about GPT-5 include its short, robotic replies, reduced personality, and the need for complex prompt engineering to achieve deeper reasoning. Users also noted increased restrictions, usage limits, and occasional bugs in complex code generation. Some felt the update was more about cost savings and efficiency improvements rather than groundbreaking advancements, as GPT-5 did not significantly enhance multimodal capabilities or creative outputs like video or audio generation. The lack of transparency in model routing and reduced user agency further frustrated users who want more control over how the AI operates.
Despite the criticisms, GPT-5 remains a powerful tool, especially for specialized tasks like frontier research and complex problem-solving, where it outperforms previous models. The video’s creator emphasizes that the mixed reactions largely depend on individual use cases and expectations. Moving forward, there may be a need for AI developers to consider creating purpose-built models tailored to specific functions or user preferences, balancing efficiency with user satisfaction to maintain broad appeal and usability.