Backlash Against OpenAI’s GPT-5 Launch Leads to Revival of GPT-4o and Introduction of New Custom Modes for Enhanced Warmth, Personality, and User Personalization in ChatGPT

Following the launch of the GPT-4 model by OpenAI, users enthusiastically embraced it, commending its enhanced accuracy and human-centric tone that fostered a more conversational atmosphere. However, frustration erupted when GPT-5 was introduced in ChatGPT without prior notice. Users quickly identified a significant shift—not only in the performance of the new model but also in its personality. Despite being marketed as the most advanced version to date, with superior reasoning capabilities, many users lamented the loss of the warmth and engagement they had appreciated in GPT-4, leading to widespread criticism.

Significant Backlash Over Abrupt Transition to GPT-5

When OpenAI officially released GPT-5 on August 7, 2025, it instantly became the default model for all logged-in ChatGPT users. This version included several tailored iterations, such as GPT-5 mini and GPT-5 nano, aimed at addressing various user needs. Unfortunately, what was meant to be a celebration of technological advancement shifted to discussions focused on performance issues and a perceived lack of personality and warmth compared to its predecessor, GPT-4. In response to the overwhelming feedback, OpenAI decided to reinstate GPT-4 as an alternate option for Plus subscribers, effectively allowing users more choices.

Moreover, OpenAI implemented several improvements in the GPT-5 model. Users are now able to select from different operational modes—Fast, Auto, and Thinking—allowing the AI to better adapt to individual needs and provide more nuanced responses. The company is committed to integrating the best features of GPT-4 into future updates, striving to combine the beloved characteristics of the older model while enhancing speed, reasoning, and overall functionality of GPT-5.

Addressing the situation, Sam Altman acknowledged the oversight and expressed the importance of personalizing user interactions. He reassured users that future iterations would evolve to maintain a warmer tone without falling into the pitfalls of being overly intrusive, as some felt the GPT-4 model could be. This incident also underscores a critical concern within the AI development community: technological advancements must prioritize user connection to avoid alienation, emphasizing the need for AI to serve its intended purpose effectively.

Updates to ChatGPT: You can now choose between “Auto”, “Fast”, and “Thinking” for GPT-5. Most users will want Auto, but the additional control will be useful for some people. Rate limits are now 3, 000 messages/week with GPT-5 Thinking, and then extra capacity on GPT-5 Thinking…

— Sam Altman (@sama) August 13, 2025

This scenario highlights a vital aspect of AI evolution: balancing innovation with continuity. As rapid advancements are made, it is crucial that these updates do not compromise user expectations and experiences. OpenAI’s approach demonstrates a thoughtful engagement with the user community, showing a willingness to adapt based on feedback and improve the technology that serves them.

Source & Images