ChatGPT Gets Smarter Memory

OpenAI is deploying a significant upgrade to ChatGPT's memory capabilities, aiming to create a more helpful and context-aware AI assistant. Codenamed 'Dreaming,' the new system is designed to overcome the limitations of previous memory functions, which struggled with staleness and accuracy over time.

StartupHub data

Companies working on this

Profiles of the companies named in this story, with funding and a one-liner from our database.

OpenAI

Private / $100B+ est

OpenAI is an AI research and deployment company dedicated to ensuring that artificial general intelligence benefits all of humanity.

Last updated: August 2026

OpenAI scores 84 out of 100 in StartupHub.ai's AI company rankings, one of the highest scores among the 13,000+ companies we track, reflecting its sustained position as the dominant force in user-facing AI products.

The company announced the rollout of a more scalable system for synthesizing memory, tackling issues like outdated information and the challenge of maintaining context across millions of users and multi-year interactions. This update is initially available to Plus and Pro users in the US, with wider availability planned.

Memory's Evolution

Launched in April 2024, the initial 'saved memories' feature allowed users to explicitly instruct ChatGPT to retain information. However, this method was akin to a user taking notes, often missing nuances and becoming stale or incorrect over time. This is a notable evolution from the initial saved memories ChatGPT offered.

A year later, OpenAI introduced the first version of 'dreaming,' enabling ChatGPT to automatically curate memories from chat history in the background. Unlike saved memories, this process synthesized information across multiple conversations, aiming for fresher, more relevant context without explicit user prompts.

While dreaming improved personalization, it wasn't sufficient as a standalone system until now. The latest architecture builds upon dreaming, creating a more capable and compute-efficient memory system. Synthesized memories are now reviewable via a summary page, where users can view, update, or instruct the AI on how to use the information.

Evaluating Memory

OpenAI evaluates memory based on three core objectives: carrying forward useful context, following user preferences and constraints, and staying current over time. The company claims the new dreaming-based system significantly improves recall of relevant facts.

For example, in a complex scenario like purchasing underwater photography gear, a user with memory enabled receives tailored recommendations based on their specific camera setup. Without memory, ChatGPT provides a generic list requiring extensive user research.

Similarly, when planning a trip, ChatGPT with memory can incorporate prior preferences, such as a user's interest in wildlife photography or preference for quiet dinners, leading to more personalized itineraries.

Frequently Asked Questions

What is ChatGPT's Dreaming memory system?

Dreaming is OpenAI's background memory synthesis system for ChatGPT. Rather than relying on user-instructed notes (saved memories), it automatically synthesizes context from your chat history across conversations. The result is a memory system that stays fresher and carries more nuanced preferences forward without requiring you to explicitly tell ChatGPT what to remember.

Is ChatGPT's advanced memory available to free users?

The initial rollout of the Dreaming-based memory upgrade was limited to Plus and Pro subscribers in the US. Free-tier users may have access to basic memory features, but the scalable synthesis system launched in 2026 is a premium feature. Availability is being expanded over time.

How do I review or delete ChatGPT's stored memories?

Synthesized memories are now visible via a summary page in ChatGPT's settings. You can view individual memories, update them, or instruct ChatGPT to remove or ignore specific information. OpenAI also gives users the option to turn memory off entirely from the same settings panel.

What came before the Dreaming memory system?

OpenAI launched "saved memories" in April 2024, which required users to manually tell ChatGPT what to remember. A first-generation Dreaming feature followed in 2025, enabling automatic memory synthesis from chat history. The 2026 upgrade is a more compute-efficient, scalable version of that second system, and is the current architecture underlying ChatGPT's memory as of mid-2026.

Can ChatGPT memory affect personalized recommendations?

Yes. OpenAI's internal examples show ChatGPT with memory delivering meaningfully different results on shopping recommendations and travel planning when it has stored context about your preferences. In one case study from the rollout, a user's camera setup and photography interests were used to filter gear recommendations without the user re-explaining their setup in each session.