Proof of Personhood

On the internet it’s hard to tell whether there’s really a human on the other end. Programs can write texts, post images and create accounts — millions of times, automatically. Proof of Personhood is the umbrella term for methods that check exactly this: is there a single, real human behind this account? The second half of that question matters just as much. It’s not only about “human or machine”, but also about making sure the same person doesn’t sign up a hundred times. The name should be taken literally: what’s being proven is personhood, not the name, age, or address.

When accounts become cheaper than opinions

Many things on the internet only work if one person has one voice. Votes, product reviews, free credit for new customers, waitlists for scarce tickets. As soon as someone can create a thousand accounts, the whole system tips over. Experts call this a Sybil attack — named after a book about a woman with many personalities. A single attacker shows up as an entire crowd.

Modern language models have made this problem considerably bigger. In the past, automated accounts gave themselves away through clumsy text. Today, programs write fluently, respond to follow-up questions, and at first glance seem genuine. Classic tests like clicking on traffic lights in images are now solved more reliably by machines than by humans.

That’s why the question is becoming increasingly relevant, both economically and politically. Platforms want to curb ad fraud, where programs generate clicks. Authorities are discussing how to protect public opinion-forming from automated campaigns. And the term comes up in the news whenever digital identity is discussed, or crypto projects that aim to distribute money to people.

From iris scans to a party with an attendance list

There is no single unified approach, but rather several competing ones. The best-known works with physical characteristics. A device captures, for example, the iris of the eye, because its pattern differs for every human. From the scan, a kind of checksum is calculated — that is, a long number. If the same person signs up a second time, the same number results and the system rejects them.

A second approach relies on social confirmation. People vouch for one another, similar to a guarantee at a bank. Whoever is confirmed by many already-confirmed people is considered genuine. Some projects hold video meetings for this at a fixed time. Because no one can sit in several groups at once, this makes duplicate accounts harder to create.

A third path uses existing identity documents: the passport is checked, but only the result “is a human” gets stored. So-called zero-knowledge proofs help here. They allow mathematical proof that a check was passed, without revealing the underlying data. No method is perfect: biometrics raise privacy questions, social networks can be infiltrated, and not everyone owns an identity document.

Worldcoin, bots, and the debate over digital IDs

The topic has become most visible through the project World, formerly Worldcoin. There, spherical devices scan the iris and then issue a digital proof. Several data protection authorities, including those in Germany and Spain, have examined the practice and in some cases banned it. The dispute centers on whether such sensitive body data may be collected for such a purpose at all.

In everyday life, you already encounter a watered-down version of this principle. When a platform demands a phone number or an ID photo for verification, that’s a crude substitute for Proof of Personhood. Discussions about the planned European identity wallet also touch on this topic, since it is meant to provide proof without full disclosure of data.

A common misconception is confusing Proof of Personhood with a regular identity check. An ID says who you are. A personhood proof ideally says only that you are human and have already been counted. This very minimalism is the goal — and at the same time the standard against which providers must be measured.