← Blog

Building a useful agent · Memory

Memory and dreams: creating the next generation of AI assistants

July 9, 2026 · 6 min read

Getting access to ChatGPT felt like having an expert you could ask anything. But it was a new expert every time. You always started from the beginning. A good colleague or friend is rarely the one who knows the most; they are the one who knows you, your work, your situation. What they have that the expert doesn't is a shared history. Memories.

Most of us have too much going on. Work threads, the school calendar, the renovation, the insurance renewal: every domain wants a piece of your attention, and the pile follows you around.

It would be nice to let someone else handle some of it. AI assistants seem like a natural fit: excellent at answering questions, great at research. But you are still the one taking the task forward. However good the answers get, it is still a tool.

You don't hand real work to a tool. You hand it to someone you trust.

To really let go, you need trust: trust that it will actually be handled. Between people, that trust grows in relationships, and relationships are made of a shared history. Memories.

That is why memory has become one of the hottest problems in AI. OpenAI is already on the third version of ChatGPT's memory system, Dreaming, and every serious lab is working on the same thing.

From chat windows to dreams

The window

The first ChatGPT, and every assistant like it, remembered exactly one thing: a window of the current chat. The window kept growing with every new model, but the rule never changed: the assistant could only remember what was inside it. Everything else was gone.

Atlas · your assistant · online
...so the flight lands at 14:20 on the 3rd.
Noted! Landing 14:20 on the 3rd.
And remember: aisle seats only, my knee is still bad.
the window
Can you book seats for the return flight?
Of course! Any seating preference?
...I literally told you this an hour ago.
The window era: everything above the window simply stopped existing.

You've felt this: the long chat that slowly loses the plot, the assistant that contradicts what you agreed on an hour ago. It wasn't being careless. The beginning of the conversation had fallen out of its world.

Keeping more of the conversation

By creating a summary of the oldest messages, more information fits in the window. Most providers use a version of this today.

Atlas · your assistant · online
the window
Summary of earlierTrip on the 3rd, flight lands 14:20. User prefers aisle seats (knee). Hotel booked, needs late checkout.
Can you book seats for the return flight?
Done: aisle seats on both legs, like on the way out.
Summaries: the gist stays in the window, the details don't.

It helps, but only for a while. As the conversation grows, summaries get summarized again, details quietly disappear, and the memory turns foggy. And the moment you notice the fog, trust starts to sink.

Memories and dreaming

So the modern systems added real memory: notes that persist across conversations. And the newest generation goes further, working on memory while you're away: a background process reads recent conversations, updates what it knows, and retires what's outdated.

Conversations
every chat, every day
Memory
persists across chats
Night pass
reads, updates, retires
the night pass writes back into memory while you're away
The memory era: conversations feed a store, and a background pass keeps it current.

In June, OpenAI shipped the third version of Dreaming, ChatGPT's memory system. The name is no coincidence, and that is the interesting part: as with so much in AI, the best model we have is the human one. Ours is modeled on it deliberately. Here is how.

Simulating human memory

Human memory is one of the best-mapped areas of psychology. The memories you can consciously call up come in two kinds: the things that happened to you, and the things you simply know.

Memories are created from experiences

Things you experienced, like the trip to Pisa, are captured by a part of your brain called the hippocampus and become episodic memories. They usually have a time, a place, and the people who were there.

Atlas · your assistant · online
Tuesday · Pisa
The line at the Baptistery was endless, we skipped it.
Fair call. The cathedral is usually quick to enter if you still want a look inside.
The leaning tower of Pisa and the cathedral on a sunny day, seen from a street with parked bikes and a crowd of visitorsIt really does lean!
Dinner at a tiny trattoria by the Arno. Best meal of the trip.
Wednesday
We climbed all the way to the top this morning. The whole city below, and the floor tilting under your feet. I actually teared up.
What a thing to see. Sounds like the high point of the trip.
Raw experience: a conversation, a photo, a moment worth keeping. Nobody said "remember this".
EpisodePisa · Tue
Skipped the Baptistery, the line was too long.
EpisodePisa · Tue
Best meal of the trip at a tiny trattoria by the Arno.
EpisodePisa · Wed
The view from the top of the tower, unforgettable.
Episodic memories: the moment, boiled down. The transcript is not what gets kept.

Try recalling yesterday's lunch conversation word for word and you'll see why: your brain drops transcripts too. It keeps moments.

Memories carry different weight

Not every moment weighs the same, and the weighing happens immediately, not later. In your brain the amygdala tags experiences with emotional significance as they happen: it's why you'll remember the view from the top forever and the Baptistery line not at all.

Episode
The view from the top of the tower, unforgettable.
Episode
Skipped the Baptistery, the line was too long.
Same trip, same day, very different weight. Importance is decided in the moment.

Sorting the day into knowledge

While you sleep, your brain replays recent experience, groups what belongs together, and stores the repeated patterns in your neocortex as things you simply know, no longer tied to any single day. Our agents run this as a nightly pass: related episodes find each other and are distilled into semantic memories, durable knowledge with links back to every moment that supports it.

Best meal: tiny trattoria by the Arno
Raved about the market lunch in Lyon, last spring
Skipped the Baptistery without regret
Semantic memory
You two prefer small local food places over famous sights.
distilled from 3 moments, across two trips
stays vivid, stays a moment →
Episode
The view from the top of the tower, unforgettable.
Dreaming: repeated patterns become knowledge. The vivid one-off stays a vivid one-off.

Notice what did not happen: the moment at the top didn't become a "fact". It doesn't need to. Some memories matter precisely as moments, and a system that understands that keeps them vivid instead of averaging them into a summary. Importance alone doesn't make knowledge; repetition does.

The right memory at the right moment

Every new message brings the right memories back. That is what makes an assistant feel like someone who knows you, not a stranger you brief from scratch each time. And it is what lets it give advice that fits you: the right suggestion for the right person, because it remembers who you are.

months later
We want to go back to Italy this year. Ideas?
the message is the cue
Recalled · semantic
You two prefer small local food places over famous sights.
Episode
The view from the top of the tower, unforgettable.
Recall: the context is the cue. A few right memories beat a thousand stored ones.

From a tool to a colleague

Remember where we started. A tool answers your questions and hands the task back to you. A colleague you trust takes the task and runs with it, because they know you, your work, and your situation. What separates the two is memory: the shared history a relationship is built on.

That is what everything above is for. Capturing experience, weighing it, sorting it into knowledge overnight, and pulling back the right piece at the right moment, so your assistant can build that history with you. Each part gets its own post, down to how we test it, including the time we caught our own test cheating. Memory is not a feature of the next assistant. It is what turns a tool into someone you can hand the job to.