OpenAI introduces 'Dreaming' memory system for ChatGPT
OpenAI announced a new memory system called 'Dreaming' for ChatGPT, designed to better retain user preferences and keep context relevant across conversations. The feature represents an architectural update to how ChatGPT manages long-term user context. Persistent, cross-session memory is an active area of development for consumer AI assistants.
Related guides (3)
Related events (8)
Memory and new controls for ChatGPT
OpenAI is testing a persistent memory feature for ChatGPT that allows the model to retain information across conversations. Users are given controls to manage what ChatGPT remembers, including the ability to view, edit, or delete stored memories. The rollout begins as a limited test before broader availability.
Introducing ChatGPT Agent
OpenAI has launched ChatGPT agent, a new capability that combines reasoning with tool use to autonomously complete multi-step tasks such as research, bookings, and presentation creation. The agent operates under user guidance, integrating thinking and acting in a unified workflow. This represents OpenAI's move to bring agentic capabilities directly into the ChatGPT product for general consumers.
Introducing ChatGPT
OpenAI announced ChatGPT, a conversational model trained to engage in dialogue, answer follow-up questions, acknowledge errors, challenge incorrect premises, and decline inappropriate requests. The model's dialogue format represented a significant step in making large language models accessible and interactive for general users. This November 2022 launch marked a pivotal moment in public AI adoption.
Custom Instructions for ChatGPT
OpenAI is rolling out a custom instructions feature for ChatGPT that allows users to set persistent preferences influencing all future conversations. The feature gives users more control over ChatGPT's tone, style, and context without needing to repeat instructions each session. This represents an incremental personalization capability rather than a core model or capability change.
Building more helpful ChatGPT experiences for everyone
OpenAI is announcing a set of ChatGPT safety and helpfulness improvements including new parental controls for teen users, routing of sensitive conversations to reasoning models, and partnerships with external experts. The update reflects OpenAI's ongoing effort to balance accessibility with safeguards across different user demographics. Routing sensitive queries to reasoning models is a notable architectural/policy decision that may affect response quality and safety outcomes.
New in ChatGPT for Business: April 2025 Updates
OpenAI published an April 2025 update for ChatGPT's business tier, highlighting four capability areas: the o3 reasoning model, image generation, enhanced memory, and internal knowledge retrieval. The announcement is framed around hands-on demos for enterprise users. This represents an incremental rollout of recently released capabilities into the business product line rather than a new model launch.
GPT-5.5 Instant: smarter, clearer, and more personalized
OpenAI has released GPT-5.5 Instant as the new default model for ChatGPT, succeeding the prior default with claims of smarter and more accurate responses, reduced hallucinations, and improved personalization controls. The announcement positions this as an incremental but meaningful update to the flagship consumer product. No architectural or training details are provided in the announcement body.
OpenAI Improves ChatGPT Mental Health Responses with Expert Collaboration
OpenAI worked with over 170 mental health experts to enhance ChatGPT's handling of sensitive conversations involving distress. The update improves the model's ability to recognize emotional distress, respond with empathy, and direct users to real-world support resources. OpenAI reports a reduction in unsafe responses of up to 80% as a result of these changes.