# How to Authenticate a Time Traveler

URL: https://www.mattsenter.com/blog/how-to-authenticate-a-time-traveler

*Authenticating a time traveler takes a secret that exists before anyone claims to be one, so I have invented a phrase I have never written down, typed, or told anyone. Backward time communication may be impossible, and Hawking’s chronology protection conjecture argues that physics may prevent it, but Gödel, Thorne, and Deutsch took closed timelike curves seriously, and preparing costs me almost nothing. The phrase works like shared-secret authentication, with the two parties separated by time instead of distance. It has to come first: a password invented after a contact cannot verify that contact, and inventing it in advance gives it an independent origin instead of a bootstrap paradox. Once a messenger uses it, the phrase is no longer secret, so the messenger should also bring a public key from whoever arranged the trip, who can later prove control of that key by signing a fresh challenge. Under a single, self-consistent timeline, one phrase is enough. I repeat mine silently every morning while brushing my teeth, and I suggest you invent yours today.*

I've come up with a secret phrase that I intend to use to authenticate anyone who claims to be a time traveler. I've never written it down, typed it, or told anyone what it is. It exists exclusively in my head, and I intend to keep it that way until the day someone in the future can demonstrate that they're connected to a message I've already received from a time traveler.

I realize that sounds ridiculous, but I've been thinking about this for a while. With advances in artificial intelligence, quantum computing, and our understanding of physics, I wonder whether we'll eventually discover some means of communicating backward through time. I'm skeptical that we'll ever physically transport human beings into the past, but transmitting information seems more plausible to me. Perhaps we'll find a way to send data, or even project thoughts backward through time. I have no idea how that would work, but I'm not willing to rule out what future generations might discover, and if it ever happens, I want to be prepared.

## Time Travel Isn't Entirely Science Fiction

Before dismissing this as nonsense, it's worth looking at some of the actual scientific research into time travel. This isn't exclusively the domain of science fiction writers and conspiracy theorists. Some of the most respected physicists and mathematicians in history have seriously studied the possibility.

In 1949, mathematician [Kurt Gödel published a solution to Einstein's field equations \[PDF\]](https://journals.aps.org/rmp/pdf/10.1103/RevModPhys.21.447) describing a rotating universe in which spacetime could contain closed timelike curves. Essentially, these are paths through spacetime that loop back into the past. Gödel's model doesn't describe the universe we actually observe, but it demonstrated that Einstein's equations don't universally prohibit such paths.

In 1988, Michael Morris, future Nobel Prize winner Kip Thorne, and Ulvi Yurtsever published [Wormholes, Time Machines, and the Weak Energy Condition \[PDF\]](https://physics.uwyo.edu/~mbrother/acs/p1446_1.pdf), examining whether traversable wormholes could theoretically be converted into time machines. Then, in 1991, quantum computing pioneer David Deutsch published [Quantum Mechanics Near Closed Timelike Lines \[PDF\]](https://gwern.net/doc/science/physics/1991-deutsch.pdf), exploring what happens when quantum systems interact with their own past.

Researchers at the University of Queensland even published an [experimental simulation of closed timelike curves \[PDF\]](https://quantum.technology/andrew/publications/2014/ncomms5145.pdf) in 2014. They simulated a quantum system interacting with an older version of itself. To be clear, they didn't actually send anything backward through time, but they were able to experimentally study some of the behavior predicted by these theoretical models.

Of course, there's substantial scientific skepticism about whether any of this could actually work. Stephen Hawking proposed his [chronology protection conjecture \[PDF\]](https://lweb.cfa.harvard.edu/~loeb/Hawking_C.pdf) in 1992, arguing that the laws of physics might prevent the formation of closed timelike curves and thereby prohibit travel into the past.

None of these theories or experiments demonstrates that backward time communication is achievable, and AI or quantum computing doesn't necessarily bring us any closer to it. Still, this is legitimate scientific research conducted by highly regarded scientists. I think it's reasonable to wonder what might become possible with sufficiently advanced technology and a better understanding of physics.

More importantly, preparing for the possibility costs me practically nothing.

## The Problem With Authenticating a Time Traveler

Imagine someone approaches me tomorrow and tells me they're from the future. They have important information that I need to know, and they insist that I believe them. How am I supposed to know they're telling the truth?

They could predict something that's going to happen, and that might actually be useful evidence. If they correctly predict tomorrow's lottery numbers or a stock's closing price an hour from now, I'd certainly be interested. I'd still have to wait for the prediction to come true, and predicting an event doesn't necessarily establish who they are or how they acquired the information.

What I really need is a piece of information that only I could know. Something I establish now that my future self can share with a time traveler, who can then bring that information back to me. That would give me a way to authenticate the message immediately, even if the communication only travels in one direction and never allows me to ask a follow-up question.

This is essentially how shared-secret authentication works in computer security. Two parties who possess the same secret can use that information to establish trust. In this case, however, the two parties are separated by time rather than distance, and the secret needs to exist before anyone contacts me.

## The Secret Has to Come First

The main reason to establish the phrase now is security. If someone contacts me tomorrow and I invent a password in response, that password does nothing to authenticate the encounter I've already experienced. I can't use a secret I haven't invented yet to verify information someone has already given me.

There's also a time travel problem worth considering, which is where things get a little weird.

Suppose someone travels back in time and tells me a secret phrase. Years later, I encounter that same person before their trip into the past, and I tell them the phrase so they'll know what to say when they meet my younger self.

Where did that phrase originate?

I learned it from the time traveler, who learned it from me, who learned it from the time traveler. Nobody ever actually invented it. The information exists in a loop without an identifiable origin.

This is called the [bootstrap paradox](https://plato.stanford.edu/entries/causation-backwards/#BooPar), also known as an ontological paradox. Philosophers and physicists have debated whether such causal loops are possible, but I'd rather avoid creating one with my password.

There's another scenario that's less obvious. Suppose someone approaches me claiming to be from the future, but doesn't tell me a secret phrase. Instead, I immediately decide that I need to invent one so I can authenticate them. I come up with a phrase right then and plan to tell that person when we meet again in the future, before they make their trip into the past.

The phrase itself has an identifiable origin because I invented it, so this isn't necessarily a bootstrap paradox in the strict sense. What prompted me to invent it, though? Someone claiming to be from the future approached me and gave me a reason to create it. The decision to create my authentication mechanism was already influenced by an event that may have originated in the future.

That doesn't make the phrase invalid, but it means the decision to create it wasn't independent of the alleged time travel encounter. More importantly, it still wouldn't authenticate the contact that prompted me to invent it.

**What I want is a secret with an independent origin, established before any alleged time traveler gives me a reason to invent it.**

That doesn't mean I've eliminated every causal loop. In fact, if this scheme ever works, the events surrounding it will necessarily form one under the single-timeline model I'm assuming. I might reveal the phrase in 2055 because I received it in 2027, while the only reason I received it in 2027 is that I revealed it in 2055. The information itself still has an independent origin because I invented it in 2026, even though the events involving that information are causally connected across time.

How do I know I'm not already part of some larger causal loop? For all I know, a time traveler is planning to contact me tomorrow, and something about my life has already been influenced by backward time communication. Perhaps even my decision to think about this problem was somehow influenced by events that haven't happened yet.

I can't completely eliminate that possibility, nor can I prove that my decision to invent the phrase is causally independent of every possible future event. I can at least establish the phrase before any encounter I'm aware of, and that seems like a reasonable precaution.

This is also why I think everyone should do this sooner rather than later. If you wait until someone claims to be from the future, you've already missed the opportunity to authenticate that initial encounter using a secret established in advance.

I've already invented mine, and as of November 12, 2026, nobody claiming to be a time traveler has contacted me.

## How to Prepare Your Own Time Travel Password

First, invent a phrase that would be extremely difficult for someone else to guess. Don't use a famous quotation, a common expression, or something easily associated with you. Make it unusual enough that someone independently coming up with the same sequence of words would be extraordinarily unlikely.

Next, and this is important, **don't write it down anywhere**. Don't type it into your phone, save it in a password manager, put it in an encrypted document, or tell ChatGPT about it. Don't even tell your spouse. The entire point is that you're the only person in the world who knows this phrase, and that the information has never been stored anywhere outside your brain.

Of course, this introduces a practical problem. How are you supposed to remember an arbitrary phrase for the next twenty, thirty, or fifty years?

I recommend repeating it silently to yourself every morning while brushing your teeth. You're already doing that every day, so associate your phrase with the routine. Eventually, remembering the phrase should become part of the habit. There's no guarantee that you'll never forget it, and over enough years you could even begin to misremember part of it without realizing it. Repeated rehearsal should help, and recognizing the correct phrase when someone tells it to you may be easier than recalling it perfectly from memory.

There's a limitation to keeping the phrase exclusively in my head, too. If future technology can somehow reconstruct memories or read information directly from my brain, the phrase might not be as private as I think it is. I'm already speculating about technology that could project thoughts backward through time, so it's fair to consider the possibility that future technology might also be able to extract them. Those aren't necessarily the same capability, but either way, my approach assumes that nobody has obtained the phrase through some other means.

A technical reader might also suggest publishing a cryptographic hash of the phrase today, giving me a timestamped commitment that anyone could verify later. That's a legitimate approach, and it would allow other people to confirm that I'd committed to the phrase before the alleged encounter. I'd need to use a sufficiently unpredictable phrase, since publishing a hash would also give attackers a way to test guesses. Personally, I'd rather avoid entering my phrase into a computer at all, even one operating offline.

That does mean this experiment is primarily for my own benefit. Someone else might believe my account, but without independently verifiable evidence, they couldn't establish that I actually knew the phrase before the alleged time traveler appeared.

I'm fine with that. I'm preparing to convince myself, not the rest of the world.

## What Happens When You Meet a Time Traveler?

Let's imagine it's 2027, and someone approaches me with an extraordinary claim. They've come from the future, or they've found a way to transmit information backward through time, and they have something to tell me.

It might involve some advanced computer system that can transmit data into the past. Maybe they've found a way to project information directly into someone's consciousness. They might even have built a physical time machine. I suspect transmitting information will be more feasible than transporting people, but that's just my intuition.

Whatever the mechanism, they tell me my exact secret phrase.

How the hell would someone know it? I've never disclosed it, and as far as I know, nobody has ever had access to that information except me. The fact that someone knows it doesn't automatically prove time travel, but it would be pretty damn compelling evidence that something extraordinary had happened.

Now, I realize that knowing my phrase isn't absolute proof of time travel. Maybe someone has developed technology capable of reading my mind, or perhaps I've accidentally revealed the phrase without realizing it. There are conceivable explanations other than backward time communication, but receiving that exact phrase without having previously disclosed it would certainly get my attention.

At this point, though, the phrase has served its purpose. Someone else knows it, which means it can no longer function as a secret exclusively known to me. If that person tells me the phrase again tomorrow, it doesn't prove anything new.

That raises another question: how do I authenticate subsequent messages, or verify whoever eventually claims responsibility for sending the original one?

## What If the Traveler and the Organizer Are Different People?

Suppose that in 2027, Person A approaches me and tells me my secret phrase. I have no idea who this person is, but the fact that they know my phrase makes me take them seriously.

Then, in 2055, I meet Person B, who explains that they've developed the technology and are planning to send someone back to 2027 to contact me. The person they're sending isn't Person B. It's Person A, who might not even be alive or available to speak with me in 2055.

How do I authenticate Person B?

In the single, self-consistent timeline I'm assuming, the phrase has already been delivered successfully. I'm not worried about whether Person B will somehow fail to send it. I'm worried about trusting the wrong person, someone who claims responsibility for an event they didn't actually arrange.

I've been thinking about this as a computer security problem, and public-key cryptography offers a way to address it.

When Person A contacts me in 2027, they don't just provide my secret phrase. They also provide a public cryptographic key belonging to whoever arranged the trip. That public key can be stored without compromising the secret, because public keys are designed to be shared.

I preserve that key along with the details of the encounter. Unlike my original phrase, there's no reason I can't write this information down or store it electronically.

Fast-forward to 2055, when Person B claims responsibility for sending Person A back to meet me. I can challenge Person B to prove that they possess the private key corresponding to the public key I received in 2027.

I generate a new, random challenge and ask Person B to digitally sign it. If the signature verifies against the public key I received decades earlier, I have evidence that Person B controls a cryptographic credential associated with that earlier encounter. I never have to learn their private key, and Person A doesn't need to be present.

This is called challenge-response authentication using a [digital signature](https://www.mattsenter.com/blog/internet-is-becoming-agent-first). Since I'm already speculating about advanced quantum computers, I'd also want the cryptography to be resistant to quantum attacks. NIST has already standardized [post-quantum digital signature algorithms \[PDF\]](https://nvlpubs.nist.gov/nistpubs/FIPS/NIST.FIPS.204.pdf), although I'd obviously want to use whatever remained secure in 2055.

Once Person B passes the challenge, I can reveal my original phrase. Person B can then give it to Person A, along with their public key, before Person A makes the trip back to 2027. From my perspective, that's simply how the information I received in 2027 got there in the first place.

The public key has another advantage. After the original encounter, my secret phrase is no longer exclusive, but the private key can continue to authenticate subsequent messages. Person B could sign additional communications, and I could verify them without needing another password or revealing any additional secrets. The phrase establishes the initial connection, and the cryptographic key maintains it.

Of course, there's no cryptographic mechanism that can guarantee Person B is trustworthy, or that their private key hasn't been [stolen or transferred](https://www.mattsenter.com/blog/microsoft-account-recovery-is-security-theater). The proof establishes control of a key associated with the earlier message, not the person's motives or an independently verified identity. It would be considerably stronger than simply taking someone's word for it.

There's also a fallback if Person A didn't bring a public key. The original encounter itself could contain information known only to me and the people responsible for arranging it. Perhaps Person A said something unusual, met me at a specific location, or described circumstances that I never disclosed afterward. If Person B can independently recount those details in 2055, that would provide additional evidence connecting them to the encounter.

It wouldn't be as strong as a cryptographic challenge, since someone might have obtained those details through another means. Person B might not know them if Person A never reported back. It's another possible way to investigate their claim, particularly if the original traveler is unavailable.

Ideally, though, I'd want the public key. It gives me a repeatable way to authenticate communications without requiring the original messenger to participate.

If my future self is reading this, make sure whoever you send back includes the public key.

## You Could Also Be the Time Traveler

The more I think about this, the more I suspect that if backward time communication ever becomes possible, the person contacting me might actually be me.

Suppose that in 2055 I discover some way to transmit information backward through time. I remember receiving an unexplained message containing my secret phrase back in 2027, and now I finally understand where it came from. I can send that same phrase to my younger self, along with whatever information I intended to convey.

How do I convince younger me that the message is legitimate? Well, I already know the phrase, so I include it in the message. When younger me receives it, he'll immediately recognize the phrase as something he invented and never disclosed.

This arrangement has an advantage over relying on another person. I don't have to give my secret to an intermediary in the future, and there's no ambiguity about whether the sender was supposed to have access to it. Future me already has the information because I've been carrying it around in my head since 2026.

I also like the idea that I could send the phrase back simply as a test. I wouldn't necessarily need to transmit some earth-shattering revelation. Just receiving the phrase would tell me that something extraordinary had happened, and I could start investigating from there.

There's a catch, though. In the single-timeline model I'm assuming, I can't reach 2055 having never received a message and then simply decide to add one to my past. If I eventually send myself a message in 2055, that message must already have been part of my experience in 2027. That's the causal loop I can't eliminate, even though the phrase itself originated independently in 2026.

There's also the possibility that I never live long enough to make the transmission. If I die before anyone learns my phrase, there's no future version of me available to disclose it. Unless someone acquires the information through some other means, the entire protocol ends with me.

I suppose that's another reason not to procrastinate if I ever invent a time machine.

## Why You Only Need One Password

Initially, I thought it might be a good idea to establish a few backup passwords, just in case someone claiming to be a time traveler turned out to be lying. The more I thought about it, the more I realized that I was approaching the problem as though time were moving forward for everyone involved.

Consider someone approaching me in 2055 and claiming they're going to travel back to 2027 to contact me. If they're telling the truth, and we're operating in a single, self-consistent timeline, that encounter in 2027 has already happened from my perspective. I would have received my secret phrase twenty-eight years earlier and spent the intervening years knowing that someone had communicated information to me from the future, or at least had somehow acquired my secret.

If I reach 2055 and nobody has ever contacted me with that phrase, why would I give it to someone who claims they're going to travel back to 2027? Their claim doesn't match my experience. I would already know that the event they're describing wasn't part of my past, at least not in the timeline I've experienced.

From the time traveler's perspective, receiving the phrase in 2055 and transmitting it to 2027 happens in that order. From my perspective, I receive the phrase in 2027 and only encounter the person who will eventually transmit it in 2055. By the time I meet them, I've already experienced the event we're discussing.

Of course, there could be multiple timelines, or perhaps changing the past creates an alternate branch of reality. I have no idea how any of that would work, and neither does anyone else. My approach assumes that backward communication is possible within a single, consistent history. If changing the past creates an entirely different timeline, then all bets are off.

Under the assumptions I'm making, I don't need to prepare for multiple failed attempts by giving my phrase to strangers who promise to use it someday. The evidence of a successful attempt should already exist in my past before I ever need to disclose it in the future.

Once that first message arrives, I have other ways to authenticate subsequent communications. The public-key approach provides a means of maintaining that relationship without repeatedly relying on a secret that is no longer secret.

That doesn't mean the same phrase can independently authenticate every unrelated traveler who might ever appear. It means I only need one phrase to establish this initial connection, after which I can rely on other credentials.

### What About Keeping a Few Extra Phrases?

There's another, more practical reason I'm not keeping a collection of secret phrases. I already think the chance of receiving even one authentic message from the future is extraordinarily low. Preparing for several unrelated first contacts, each requiring its own unused secret, feels like planning for an even more remote possibility.

If those encounters were independent, multiple successes could be astronomically less likely than one. For a simple illustration, if each of two independent opportunities had a one-in-a-million chance of succeeding, the chance that both succeeded would be one in a trillion. Those aren't estimates of the odds of time travel. They're just an illustration of why I don't feel compelled to memorize a whole inventory of passwords for independent encounters.

There's an important catch: the encounters might not be independent. If someone figures out how to send a message backward through time, sending a second one might be easy. Receiving that first message would also radically change my view of what's possible. I can't assume that a second message would remain extraordinarily unlikely after the first one arrives.

That's why the distinction matters. More messages from an established connection can use the public key. A genuinely unrelated first contact might call for a fresh secret, but preparing for several of those in advance adds something else to remember for a possibility I already consider remote. One phrase seems like a reasonable amount of preparation.

I'm sticking with one phrase. I just need to remember it.

## Why I'm Actually Doing This

I recognize that there's an exceptionally good chance none of this will ever matter. We don't know whether backward time communication is possible, and there are serious scientific arguments suggesting that the laws of physics may prevent it entirely, but I'm also not convinced that we understand enough about the universe to categorically dismiss every conceivable method of communicating with the past. A century ago, the technology we have today would have seemed pretty extraordinary, and I don't pretend to know what another century of scientific discovery might bring.

What interests me about this particular exercise is that there's practically no cost to preparing. I don't need to spend any money, learn anything new, or change my routine. I just need to memorize an unusual phrase and make sure I don't disclose it prematurely.

If time travel never happens, I've spent a few seconds every morning thinking about something that ultimately didn't matter. I can live with that, but if someone ever does figure out how to transmit information backward through time, I want to have already established a means of authenticating the first contact, and if it's my future self doing the transmitting, I want him to know that younger me is going to be skeptical and that simply claiming to be from the future isn't going to cut it.

The important thing is that the phrase exists before anyone contacts me. That gives it an independent origin, allows me to recognize it immediately, and avoids having to invent an authentication method in response to someone I haven't authenticated yet.

I've already done it. I have my phrase, and I haven't shared it with anyone. It's not even encoded in this blog post as an easter egg.

If you're interested in preparing for the same possibility, I suggest coming up with yours today. Don't write it down, don't tell anyone, and repeat it to yourself every morning while brushing your teeth.

For all you know, your future self is already trying to figure out how to contact you. Might as well be ready.