Tell someone their work is protected the moment they make it, and most people relax and stop thinking about it. That reaction is reasonable. It is also where the confusion starts, because "protected" and "provable" are not the same word, and the gap between them is what actually decides a dispute.
What the Berne Convention actually says
The Berne Convention is the treaty behind the rule most people have heard in passing: copyright arises automatically the moment a work is created and fixed in some form. No application, no fee, no office to file with, no waiting period. The right exists as soon as the work does.
Berne has 181 member states, which in practice means nearly every country with a functioning legal system honors the same principle. A short film you shoot in Geneva is protected in Brazil the same way it is protected in Switzerland, without you doing anything extra to make that true in Brazil. That is the treaty's actual achievement: it removed the need for a creator to register separately in every country where their work might one day matter.
Berne covers more than "you own it"
The convention does not stop at economic rights. It also protects moral rights, an author's right to be identified as the creator of a work and to object to a distortion of it, independent of who might hold the commercial rights at any given point. That protection is automatic too. It still depends on being able to show, when it matters, that you are in fact the person being named as the author, which pulls the question back to evidence rather than entitlement.
The right is automatic. The proof is not.
Berne gives you a right, not a record. Nothing in the automatic-protection rule produces evidence of what you made, when you made it, or that you were the one who made it. The right attaches silently, at the moment of creation, and by design it asks nothing of you for it to exist. That is exactly why it leaves nothing behind.
In the ordinary case this is not a problem, because nobody asks. You write, you publish, you move on, and the automatic right sits quietly underneath everything you did, doing its job without anyone ever needing to see it work.
The case where it does become a problem
A dispute almost never argues about whether copyright exists in a piece of work. Both sides usually agree that whoever created it owns it. What they disagree about is who created it, and who created it first.
Two studios pitch a near-identical concept to the same client a week apart. A freelancer's deliverable turns up, lightly modified, inside a competitor's campaign. A former collaborator launches a product with a look that is uncomfortably familiar. Nobody in these situations is confused about the law. Both sides are confident they are right, and the outcome depends entirely on which side can show a dated, verifiable link between themselves and the work, not on which side understands copyright better.
This is the point where the automatic right has already done everything it is going to do for you. From here, the argument runs on evidence, not entitlement, and evidence is not something Berne hands you along with the right.
What "proof" needs to actually hold up
Proof that survives being challenged needs three things at once, and most of what people reach for in the moment has one or two, not all three.
It needs to be tied to the exact file, not to a title, a description, or a folder name, because disputes are almost always about which version came first. It needs a time source you do not control, because a date you set yourself is a date you will be asked to justify, and asking someone to trust your own clock is not evidence. And it needs to be checkable by someone else without your cooperation, because the person you are trying to convince is, by definition, the person who has already decided not to take your word for it.
A file sitting on your laptop meets none of these. A screenshot of a creation date meets none of them either, because that date lives on a system you control and can be altered.
Where the common shortcuts fall short
Emailing the file to yourself. The instinct behind it is correct: get a dated mark on the work from somewhere outside your own control. The weakness is that an email header can be edited on either end, and an unopened message proves less than people assume once it is actually challenged in a real argument.
Relying on a platform's upload date. Cloud storage, a CMS, a design tool. The date sits with a company you do not control, under terms that can change, in an account that can be closed or lost. It records when you uploaded, not when you created, and it is only as durable as that platform's willingness to keep the record and hand it over when asked.
Doing nothing and pointing at Berne. Technically correct and practically the weakest position of all, because it answers a question nobody in a real dispute is asking. Nobody doubts that copyright exists somewhere in the work. They doubt that it is yours, or that it is yours first.
What actually closes the gap
The fix is not a second registration system layered on top of Berne, and it does not need to be one. What closes the gap is producing, at the moment of creation, a record that meets the three requirements above: bound to the exact file, timestamped by a source you do not control, and independently checkable by someone who has never met you.
A qualified electronic timestamp does that. Under eIDAS, a qualified timestamp issued by a Qualified Trust Service Provider carries a legal presumption of the accuracy of the date and time it records, and of the integrity of the data it is bound to. Swiss Trust Layer issues these through Swisscom Trust Services, a ZertES-accredited provider, so the timestamp on a sealed file is not a claim we make about our own system. It is a qualified one under Swiss and EU law.
The seal binds to a cryptographic hash of the exact file, not to a description of it, so the "which version" question that most disputes turn on has a direct answer built in from the start. We also register that content hash, generated as an ISCC, in a global public database, which makes the content itself traceable for infringement and AI-training-licensing purposes without ever exposing the underlying file to anyone who looks it up.
None of this creates the copyright. Berne already did that, automatically, the moment the work existed. What it creates is the part Berne was never designed to produce: a dated, independently verifiable answer to "prove it," ready before anyone asks the question, instead of assembled in a hurry after they do.
The part worth remembering
Automatic protection and provable authorship solve two different problems, and it is easy to assume that having one means you have the other. You do not need to distrust Berne to notice the gap between them. You need to notice that the gap is invisible right up until the day someone disputes what you made and when, and by then it is too late to build the record you needed on day one. See how sealing and timestamping actually work for the mechanics, or read how the same evidence question plays out from the other direction in our Article 50 coverage.
