According to Sam Altman (@sama) on Twitter, OpenAI plans to relax some of the current restrictions on ChatGPT, aiming to address user feedback that the platform had become less engaging due to strict content limitations related to mental health concerns. Altman stated that, after developing tools that better mitigate mental health risks, OpenAI will introduce a new version of ChatGPT within weeks that allows for more expressive, human-like AI personalities, including more natural conversational styles and customizable behaviors. This update is designed to enhance user engagement and satisfaction, particularly for business applications that benefit from personalized AI interactions. Additionally, OpenAI announced plans to implement full age-gating by December 2025, enabling adult users to access more mature content, such as AI-generated erotica, for verified adults. This move reflects a broader industry trend toward offering tailored AI experiences to different user segments, which could open new business opportunities in areas such as digital companionship, entertainment, and adult content. Source: Sam Altman (@sama), Twitter, Oct 14, 2025.

According to Sam Altman (@sama), ChatGPT has introduced new user-selectable modes for GPT-5: 'Auto', 'Fast', and 'Thinking'. This feature enables users to tailor their AI experience based on speed and depth of reasoning, with 'Auto' as the default and additional granular control for advanced users. The update also raises rate limits to 3,000 messages per week for the 'Thinking' mode, with extra capacity available, significantly enhancing productivity and scalability for enterprise and power users. These enhancements present new business opportunities for AI-driven workflows, allowing organizations to optimize response times and computational resources for varied use cases (source: Sam Altman, Twitter, August 13, 2025).

According to DeepLearning.AI, the instructors of the 'Reinforcement Fine-Tuning LLMs with GRPO' course are hosting a live AMA to discuss practical applications of reinforcement fine-tuning in large language models (LLMs). The session aims to provide real-world insights on how Generalized Reward Policy Optimization (GRPO) can be leveraged to enhance LLM performance, improve response accuracy, and optimize models for specific business objectives. This live AMA presents a valuable opportunity for AI professionals and businesses to learn about advanced methods for customizing AI solutions, ultimately enabling the deployment of more adaptive and efficient AI systems in industries such as finance, healthcare, and customer service (source: DeepLearning.AI Twitter, June 13, 2025).