Skip to content
Photo evidence7 min read

How to Prove When a Photo Was Taken

To prove when a photo was taken, do not rely on the phone. Capture the photo live and have an independent timestamp authority certify, at that moment, that the file existed. The date then no longer depends on either party.

By Cédric Driaux, Founder of tryVidimus

Drawing of a photo with a clock

Why the phone's date isn't enough

Every photo carries dates. None of the usual ones can prove when a photo was taken, because each is written by a device or a person with an interest in the answer. This is one of the weak points listed in our photo evidence guide.

Date you seeWho writes itWhy it proves little
Date in the photo's metadata (EXIF)The phone, from its own clockThe clock can be changed in the settings, and the field can be edited afterwards.
File creation or modification dateThe operating systemIt can change when the file is copied, downloaded or saved again.
Date printed on the imageAn app, or anyone with an editorIt is part of the picture. A date can be drawn on any image.
Date of the message or email that carried itThe messaging or email serviceIt shows when the file was sent, not when the scene was photographed.
Date recorded by the platform that stores itOne of the parties, or its softwareThe other side has to trust the party that keeps the record.

The first line is the one people trust most, and wrongly. A separate article explains why EXIF metadata can be faked and what it remains useful for.

Take a rental agency. A car comes back on Friday evening, and on Tuesday the agency sends the customer a photo of a cracked bumper. The customer asks one question: was this photo taken on Friday, or on Monday after someone else moved the car?

The agency points to the date in the file. The customer answers that the agency controls both the phone and the file. Both are right, and the date settles nothing.

Independent timestamp authorities

The way out is to let someone else write the date: a third party with no stake in the dispute, whose only job is to certify time. That is a timestamp authority.

The protocol is public. It is defined by RFC 3161, the Internet standard for time stamps, and it works in three steps.

  1. A fingerprint of the data is computed, here a SHA-256 hash. The fingerprint identifies the data without revealing it.
  2. The fingerprint is sent to the authority. The authority never receives the photos themselves.
  3. The authority returns a signed token that binds this fingerprint to the date and time on its own clock.

The token proves one precise thing: data with this fingerprint existed at that date. The token cannot exist before the data does, and nobody can change the date without breaking the authority's signature.

The same fingerprint also protects the content. If the file changes later, its fingerprint no longer matches the token, as described in how to prove a photo has not been edited.

With both together, the window closes. The photo cannot be older than the capture session, because it came from the camera during that session. It cannot be newer than the token, because the authority signed its fingerprint.

Qualified timestamps under eIDAS

Not all timestamps have the same standing. In the European Union, the eIDAS Regulation distinguishes ordinary electronic time stamps from qualified ones, issued by a qualified trust service provider.

Article 41 of the eIDAS Regulation sets two rules. An electronic time stamp cannot be denied legal effect or admissibility as evidence solely because it is in electronic form. A qualified electronic time stamp enjoys a presumption of the accuracy of the date and time it indicates, and of the integrity of the data to which they are bound.

A presumption is a starting point, not a final word. The date is taken as accurate unless someone shows otherwise, and the other side remains free to try.

The presumption covers the date and the integrity of the data. It says nothing about what the photo shows. Our article on qualified electronic timestamps under eIDAS goes through the text in more detail.

tryVidimus requests one qualified electronic timestamp per proof. In production it is issued by Certigna, a French qualified trust service provider. The time shown as "captured at" is different: it is the time observed by our service, and it is not qualified.

How to check a timestamp yourself

A timestamp is only worth something if the other side can check it without trusting you. Three checks are enough, and none of them needs our help.

  1. Check that the token covers this proof. The fingerprint inside the token must match the fingerprint of the proof you are looking at.
  2. Check the signature. The token is a standard RFC 3161 token, so OpenSSL can verify that the authority signed it and that it has not been altered.
  3. Check who signed. The European Commission publishes the trusted list of qualified providers, where you can confirm that the authority is qualified for time stamps.

For someone who is not technical, each proof has a public verification page, open without an account. For someone who wants to rerun everything, the open source verifier performs the checks on their own machine.

Back to the agency. With a sealed proof captured at the return desk on Friday, the answer to the customer's question is a link. The customer opens it, sees the certified date, and can have anyone verify the token.

The discussion that remains is the real one: who is responsible for the crack. The proof does not answer that, and it is not meant to.

Frequently asked questions

Does the date in a photo's metadata prove when it was taken?

No. The date in the metadata is written by the phone from its own clock, and that clock can be changed in the settings. The field itself can also be edited afterwards with common tools. It is useful context, for sorting or for spotting inconsistencies, but it does not establish a date against someone who contests it.

Can I timestamp a photo that already exists?

Technically yes, but it proves less than people expect. The timestamp shows that the file existed when it was stamped, not when the scene was photographed. An old photo stamped today gets today's date. To prove when a photo was taken, the capture has to be live and the proof has to be stamped when it is sealed.

What is a qualified electronic timestamp?

It is an electronic time stamp issued by a qualified trust service provider under the European eIDAS Regulation. Under Article 41 of that regulation, it enjoys a presumption of the accuracy of the date and time it indicates and of the integrity of the data bound to it. That presumption applies in the European Union. This is not legal advice.

Is the timestamp useful outside the European Union?

Yes, as a technical fact, though without the eIDAS presumption. Outside the Union, the token remains a signed record from an independent authority, which anyone can verify with standard tools. How much weight a court or an insurer gives it depends on local rules, so ask a lawyer about a specific case.

Does the timestamp authority receive my photos?

No. The authority receives the fingerprint of the proof, never the photos. A fingerprint identifies the data without revealing it, and the image cannot be rebuilt from it. The authority returns a signed token that binds the fingerprint to a certified date, and that token can be checked later by anyone.

Is the "captured at" time the certified time?

No. The time shown as "captured at" is the one observed by the service when the photo was taken, and it is not qualified. The certified time is the one in the timestamp token, issued by the authority when the proof is sealed. Only that second time carries the presumption described in the eIDAS Regulation.

Try it in the sandbox

You cannot prove when a photo was taken with a date that one of the parties controls. A live capture followed by an independent timestamp gives a date that anyone can verify. You can try the full journey for free in the sandbox, without a card.

Create an account

Official sources

Cédric Driaux · Founder of tryVidimus

Cédric Driaux is the founder of tryVidimus, published by Fyher. He has been a software developer for 23 years. He also builds ScanStay, a photo proof tool for short-term rental hosts.

Related articles