Is AI intelligence at its peak? Sam Altman: The core of GPT-6 will focus on "memory" and "personalization", without emphasizing being smarter.

OpenAI announced that GPT-6 will shift towards "memory" and personalization in response to the GPT-5 controversy and the new ideological regulations of the Trump administration, alongside emerging privacy concerns and a technological ceiling. (Background: A comprehensive summary of OpenAI's strongest release, GPT-5: free access, feature highlights, gpt-5, gpt-5-mini, and gpt-5-nano API pricing.) On the 8th of this month, OpenAI officially launched the latest model, GPT-5, but the community had mixed reactions to the model. Some believe it has become smarter, while many others complain that the results generated by GPT-5 are worse than before, and it clearly has not reached the level of AGI claimed by Sam Altman. This led the community to discuss whether the intelligence of LLM models, trained on nearly all human data, has already peaked. In this context, Altman talked about GPT-6. Recently, OpenAI CEO Sam Altman revealed signs of this trend during an interview. When he mentioned the next generation large language model, GPT-6, he stated that the team's focus will be on "memory" and "personalization," rather than emphasizing how much smarter the new generation of models will be compared to previous ones. Memory is key to truly personalizing ChatGPT. It needs to remember who you are—your preferences, habits, and quirks—and adjust accordingly. After receiving criticism from users for being less intelligent and more indifferent in GPT-5, this shift is particularly notable. Although it indicates that OpenAI intends to reposition its product from being a "universal answerer" to a "long-term conversational partner," hoping to enhance user engagement through memory functionality, it also seems to acknowledge that it is becoming increasingly difficult for the market to experience the same leaps in intelligence as seen in previous generations. The Trump administration's executive order and ideological neutrality is another factor. The Trump administration issued an executive order in July requiring that AI systems purchased by the federal government maintain "ideological neutrality" and support customization. In response, Altman indicated that GPT-6 will adopt a "centrist" default while allowing flexibility for users to decide the tone. He stated: I believe our products should have a more moderate, centrist position, and then you can push it quite far. Even language models must align with different stances, showing that AI has transitioned from pure technical achievements to a public infrastructure that faces diverse values. Technological ceilings and capital pathways While we all hope that AI can continue to grow exponentially, perhaps we should start thinking about where the next growth will come from as large language models slow down in terms of intellectual augmentation. OpenAI bets on "memory" and "personalization" while exploring peripheral areas like neural interfaces, energy, and robots. Market funding may also need to consider shifting the computing power competition towards investing in AI application layers that are more closely aligned with human needs (though competition is becoming fiercer; think about the internet bubble era, where over 95% of internet applications did not survive...). In summary, the information disclosed by Altman suggests that GPT-6 seems to prioritize "understanding you better" over being "smarter." How funds, enterprises, and you—who are undergoing AI transformation—consider the next steps will be key points to observe. Related reports: Is your TikTok Shop still safe? AI + fake apps + phishing pages, the new scam "ClickTok" is targeting your crypto wallet. How long has it been since you seriously thought about it? Princeton scholars warn: AI is eroding the "heritage of the Enlightenment." Lawyer Lin Shanglun's special article: Did AI defeat three lawyers? Don't get it wrong, this is the prologue of "Lawyers 2.0." <Is AI intelligence at its peak? Sam Altman: The core of GPT-6 will focus on "memory" and "personalization," without emphasizing being smarter.> This article was first published on BlockTempo, the most influential blockchain news media.