How AI Can Help You Remember What You Forgot
- Pablo Vazquez
- Aug 8
- 11 min read
You’re walking out the door when someone texts:
“Can you send me that file later?”
You read it.
You think:
Yeah, I’ll do that.
Then you leave.
Six hours later, you remember.
Or maybe you don’t.
This happens constantly.
Call the dentist.
Reply to that email.
Bring something home.
Send the link.
Cancel the subscription.
Look into that restaurant.
Follow up with someone next week.
Buy the thing before the trip.
None of these tasks are particularly difficult.
The hard part is remembering that they exist at the right moment.
Psychologists have a name for part of this: prospective memory, our ability to remember that we intended to do something in the future. It covers ordinary situations like remembering to send a message later, attend an appointment, or complete something after another event happens.
And modern life gives that memory system a lot to keep track of.
Your intentions are being created inside texts, emails, conversations, calendars, meetings, apps and random thoughts throughout the day.
AI creates an interesting possibility:
Instead of relying entirely on you to remember everything, technology can start helping you keep track of the things you meant to do.
Why Do We Forget Small Tasks?
It can feel ridiculous when you forget something simple.
You remembered a huge presentation.
You remembered your flight.
But somehow you forgot to put the laundry in the dryer.
The problem is often not that the task was too difficult to remember.
It’s that remembering to do something later has to compete with whatever else your brain is already doing.
Research on prospective memory has found that remembering future intentions can become harder when people are simultaneously dealing with demanding ongoing tasks, distractions, and competing mental load.
Think about a normal moment:
You’re answering an email.
A Slack notification appears.
Someone texts:
“Grab milk on the way home.”
You answer:
“Yep.”
Then go back to the email.
There was never a moment where you sat down and created a formal task called:
Buy milk.
You simply formed an intention.
And then life continued.
A Lot of What We Forget Starts With “I’ll Do That Later”
This phrase is dangerous:
I’ll remember.
Because when you say that, you’re basically asking your brain to interrupt you at exactly the right future moment.
Sometimes it does.
Sometimes you're standing in your kitchen at 11:43 PM thinking:
Oh shit. I was supposed to do that.
A future task generally needs some kind of cue.
Maybe:
At 3 PM, call the doctor.
Or:
When I get home, take the package inside.
Research distinguishes between intentions triggered by a time and intentions triggered by an event or situation, and context can play an important role in helping the intended action come back to mind.
That's why you can genuinely intend to do something and still completely miss it.
The intention existed.
The right reminder didn't.
The Problem With Relying on Your Brain as a Task Manager
Your brain is great at a lot of things.
Being the permanent storage system for every tiny promise you make probably shouldn't be one of them.
Imagine trying to mentally maintain:
Call Dad tonight.
Reply to Sarah tomorrow.
Book dentist this week.
Bring charger Wednesday.
Look at hotel options.
Send deck before meeting.
Cancel trial before renewal.
Ask professor about assignment.
Pick up dog food.
Follow up with Alex.
You can probably hold some of those.
But now add everything else you’re thinking about.
Work.
Money.
Relationships.
School.
What you’re eating tonight.
Where you parked.
What happened yesterday.
What you’re trying to accomplish next month.
Eventually something drops.
That doesn't necessarily mean you need to “try harder to remember.”
It may mean you need a better place for those intentions to live.
Reminders Help, But Only When You Remember to Create Them
Reminders are one of the obvious solutions.
And they work.
Research has repeatedly found that external reminders can improve people's ability to carry out future intentions, particularly when memory demands increase. A recent study found that using reminders could substantially reduce age-related differences in prospective-memory performance under demanding conditions.
The problem is one step earlier.
To set a reminder, you first have to recognize:
This is something I need to remember.
Then you have to:
Open the reminders app.
Create the task.
Pick the right time.
Give it enough information.
And hope that the time you selected actually makes sense when the reminder appears.
That's fine for:
Dentist appointment tomorrow at 9.
It's less perfect for:
Follow up with Chris after he sends me the proposal.
What time should that reminder be?
You don't know.
The useful cue is not necessarily Tuesday at 2:15 PM.
The useful cue is:
Chris sent the proposal.
That's where context starts mattering.
Why Reminders Fail When Your Context Is Scattered
Imagine you set:
Reminder: Book hotel Friday.
Friday arrives.
Your phone says:
Book hotel.
Okay.
Hotel for what?
Which city?
What dates?
What was your budget?
Who are you traveling with?
Where was the event again?
Now you're searching email.
Then messages.
Then your calendar.
Then maybe Notes.
The reminder successfully remembered the task.
But you still have to reconstruct the context.
This is a major limitation of a lot of digital organization.
The reminder exists in one place.
The information needed to act on it exists somewhere else.
Your calendar knows the dates.
Your email has the confirmation.
Your friend sent the location.
Your Notes app has the hotels you liked.
Your brain is responsible for connecting everything.
That's not really a memory system.
That's a scavenger hunt.
Some Reminders Also Arrive at the Wrong Moment
Consider:
Call pharmacy
Your phone reminds you at 8 AM.
You dismiss it because you're driving.
Now what?
Maybe you'll remember again.
Maybe not.
Research on prospective memory suggests that context and the predictability of relevant cues affect when people allocate attention toward remembering intended actions. In plain English: a reminder is more useful when it shows up in a situation where the action actually makes sense.
That's an important idea for the future of reminders.
Instead of only:
Remind me at 3 PM.
technology can become better at:
Remind me when this becomes actionable.
This Is Where AI Starts Getting Interesting
Traditional reminders depend heavily on you.
You create them.
You word them.
You schedule them.
AI potentially lets us reverse that relationship.
Instead of asking:
What did I remember to create a reminder for?
AI can help ask:
What happened today that seems like something I need to remember?
Imagine AI is allowed to use context from the apps you choose to connect.
It sees:
Text: “Can you send me those photos tonight?”
That's probably an action.
It sees:
Email: “Your application is incomplete. Please upload the missing document before Monday.”
Definitely an action.
It sees:
Calendar: Flight Thursday morning.
And an email:
Online check-in opens Wednesday.
That creates another useful future action.
AI is not magically reading your mind.
It's looking for clues inside information that already exists.
How AI Can Detect “Open Loops”
“Open loop” is useful everyday language for something that still feels unfinished.
It isn't one single, precise scientific category, but the idea overlaps with research on unfinished intentions and prospective memory: things we intend to complete later can continue to require attention until they're completed, abandoned, or put into a reliable plan.
An open loop might be:
Something you promised
“I'll send that tonight.”
Something someone asked for
“Can you review this?”
Something waiting on another event
“Once she replies, book the meeting.”
Something approaching
Subscription renews Friday.
Something incomplete
Application missing one document.
Something changed
Appointment moved from Monday to Wednesday.
This is exactly where AI can add value.
Instead of creating 40 reminders, it can potentially identify:
These few things still seem unresolved.
The Goal Is Not for AI to Remind You About Everything
This part is critical.
Imagine your AI noticed every possible unfinished thing.
You haven't answered this person.
You haven't read this article.
You haven't finished that Netflix episode.
You haven't booked the restaurant.
You haven't responded to this promotional email.
Congratulations.
Your AI is now the world's most annoying parent.
A useful memory system requires judgment.
Some things matter.
Some things expired.
Some things were casual conversation.
Some things were already handled.
Some things can safely disappear.
The important question isn't:
Can AI detect unfinished things?
It's:
Can AI determine which unfinished things actually deserve to come back to you?
Context Helps AI Decide What Is Worth Remembering
Imagine three messages:
“We should check out that coffee place sometime.”
“Can you send me the contract today?”
“Your credit card payment is due tomorrow.”
All three could technically become tasks.
But they aren't equally important.
Context helps separate them.
The contract has a clear commitment.
The payment has a deadline.
The coffee shop is an idea.
Now imagine AI also knows:
You already sent the contract.
Your credit card is on autopay.
Suddenly neither needs to bother you.
That's what a better memory system looks like.
Not remembering everything.
Remembering what still matters.
AI Can Help You Remember Before You Realize You Forgot
This is the real shift.
Most reminder systems work like this:
You remember → you create a reminder → technology reminds you later.
But there's a hole in that system.
What happens when you never create the reminder?
AI can potentially help fill that gap:
Something happens → AI recognizes a possible commitment → AI keeps track → it surfaces it when relevant.
That's much closer to how a human assistant works.
A good assistant doesn't only remind you about things you formally entered into a database.
They might say:
“You told James you'd send that today.”
Or:
“You said you wanted to book that before Friday.”
Or:
“They replied. You can move forward now.”
That's not just a better alarm clock.
That's continuity.
How to Build a Better Personal Memory System Today
You don't have to wait for perfect personal AI.
You can already make your existing system much better.
The basic idea is to stop making your brain responsible for storage.
A good personal memory system should do five things:
Capture: Get commitments out of your head quickly.
Preserve context: Keep enough information to understand the task later.
Surface it at the right time: Use deadlines or useful events rather than random reminders.
Prioritize: Don't treat every possible task as equally important.
Close the loop: Complete it, postpone it intentionally, delegate it, or remove it.
The goal isn't to remember more.
It's to need your memory less.
Step 1: Capture the Intention Immediately
When you think:
I need to remember this later...
that is the moment to capture it.
Not ten minutes later.
Not tonight.
Now.
It can be simple:
Remind me to email Chris after lunch.
Add buy passport photos to tomorrow.
Remember I promised Mom I'd call tonight.
Whatever system you use, the key is getting the intention somewhere outside your head.
Step 2: Keep the Why With the What
Don't save:
Call Sarah
Save:
Call Sarah about Friday's venue before she confirms the reservation.
The second version contains enough context to be useful when you see it tomorrow.
This also makes future AI dramatically more useful.
An AI can prioritize:
Call Sarah before reservation is confirmed.
much better than:
Call Sarah.
Step 3: Use the Right Cue
Ask:
When will this actually be useful?
Maybe it's a time.
Pay rent on the 1st.
Maybe it's a place.
Buy batteries when I'm at the store.
Maybe it's another event.
Follow up once Alex sends pricing.
Research on prospective memory has long distinguished between these time-based and event-based cues, and the surrounding context can affect whether the intended action comes back to mind successfully.
Step 4: Review What Is Still Open
Once a day—or whenever your system can do it for you—look at:
What am I still waiting on?
What did I promise?
What's approaching?
What can I remove?
This isn't about building another giant list.
It is about preventing small unfinished things from quietly disappearing.
Step 5: Let AI Do More of the Sorting
This is where the system can get lighter.
You can give an AI assistant:
Messages.
Meeting notes.
Emails.
Your calendar.
A messy task dump.
And ask:
“What commitments, deadlines, follow-ups or things I said I'd do are hiding in here?”
Then:
“Which of these still appear unresolved?”
Then:
“Which three actually deserve my attention?”
The important progression is:
Find → filter → prioritize.
Not:
Find → dump 72 new tasks on me.
You Don't Need Perfect Memory
That's the larger idea.
People often treat forgetfulness like a personal failure.
But we use external memory systems constantly.
Calendars.
Alarms.
Sticky notes.
Lists.
Contacts.
Photos.
Bookmarks.
Reminders.
The question isn't whether technology should help us remember.
It already does.
The interesting question is:
How much better can that help become when technology understands context?
A reminder knows:
Do X.
A context-aware AI can potentially understand:
You said you'd do X, it hasn't happened yet, someone is waiting for it, and now is a good time to handle it.
That's a meaningful difference.
Where NOA Fits
This is almost exactly the problem NOA is being designed around.
NOA's internal product direction centers on people who have commitments and information distributed across communication, calendar, productivity, travel and other apps—and specifically calls out staying ahead of deadlines, follow-ups and changes before they are missed.
Rather than requiring someone to remember every possible task and manually enter it, NOA is being built to use context from the apps they choose to connect, alongside plans and priorities, to turn that information into personalized actions.
That's an important distinction.
The goal isn't:
Remember everything yourself.
And it isn't:
Get reminded about everything constantly.
It's closer to:
Let the technology keep track, then bring back the things that actually deserve your attention.
NOA's product model also separates that into familiar human questions: what needs to get done, what's ahead today, and what you may want to ask or delegate.
Because sometimes the thing holding you back isn't a huge decision.
It's the tiny thing you forgot three days ago.
Frequently Asked Questions
Why do I keep forgetting small tasks?
Small tasks often depend on prospective memory: remembering an intention at the right moment in the future. That can become harder when you're simultaneously busy, distracted, or managing other demanding tasks.
What is prospective memory?
Prospective memory is the ability to remember to perform an intended action later—for example, remembering to call someone, attend an appointment, or send something after another event occurs.
Do reminders actually help memory?
Yes. Research supports using external reminders to improve prospective-memory performance, particularly when people are carrying multiple intentions or dealing with greater memory demands.
Why do phone reminders sometimes not work for me?
A reminder can appear when the task isn't actionable, contains too little context, or simply be dismissed while you're focused on something else. Research on prospective memory suggests that contextual cues and timing can influence how successfully future intentions are remembered.
Can AI remind me about things I forgot?
AI can help identify possible commitments, deadlines, and follow-ups from information you provide or from connected sources the AI has permission to access.
The more advanced idea is not simply setting an alarm, but identifying that an unfinished action exists before you've manually created a reminder for it.
What is an open loop?
In everyday productivity language, an open loop is something unfinished that may still require attention, such as a promise, unanswered request, approaching deadline or task waiting on another event.
It overlaps with psychological research on prospective memory and unfinished intentions, though “open loop” itself is a broader productivity term rather than one precise scientific construct.
How can I build a better memory system?
Capture commitments quickly, preserve the context around them, use reminders tied to useful times or events, regularly clear things that no longer matter, and use AI to help identify and prioritize unresolved actions.
A good memory system reduces how many future intentions you have to actively carry in your head.
The Bottom Line
Most things we forget aren't forgotten because they're unimportant.
They're forgotten because they showed up in the middle of something else.
A text.
An email.
A conversation.
A thought while walking somewhere.
A promise you made quickly.
Then life kept moving.
Traditional reminders help, but only after you realize something needs to be remembered.
AI creates the possibility of something better:
Technology that helps notice what you meant to do, keeps enough context to understand it, and brings it back when it actually matters.
Because a better memory system shouldn't require you to become better at remembering everything.
It should make remembering everything less necessary.
Comments