The Internet was built without a way to know who and what you are connecting to…. There is no consistent and comprehensible framework allowing [users] to evaluate the authenticity of the sites they visit, and they don’t have a reliable way of knowing when they are disclosing private information to illegitimate parties.
— Kim Cameron, Architect of Identity, Microsoft Corporation
The Internet as it currently exists has no mechanism for securing the identity of its users. This affects our ability to know with whom we are interacting and sharing, to what degree and level of privacy, and, increasingly, to what end.
The problem is not that the Internet is anonymous but that it has become both pathologically pseudonymous and a place where a tremendous financial incentive has emerged to perpetuate an intrusive model of identity that is unnecessarily restrictive and ultimately dehumanizing. Users can pretend to be anyone, to create new identities for themselves, to impersonate others, to avoid accountability; while artificial intelligence of various kinds can present itself as being human with an astonishing rate of success.
At the same time, human Internet users are increasingly monetized and surveilled. Because most of us do not understand how valuable our identities are — our personal data, our preferences — we have gradually become the product of unaccountable corporate enterprises that have no desire to limit what they take or exercise caution in how they use it. Too many otherwise secure online services demand access to private information that they do not require; this information is stockpiled and invariably used against the better interests of the individual, which includes our communities and humanity in aggregate. Thus, our digital identity has been diverted to accommodate the needs of our worst inclinations. We stand ready to repeat the same mistakes that have led to oppression, inequality, and tyranny.
Yet we do not accept this as our end state. These caustic factors are inherently regressive in a world that has become aware of itself on a global scale for perhaps the first time in history and that is using this awareness to improve its condition. New technologies to supercharge this change, from decentralized economies to direct digital democracies, are ready to deploy but for the identity challenge.
This article sketches the contours of the problem and offers what we believe is the best way forward: Idena’s proof-of-person blockchain.
The identity problem
Any system that formalizes ownership and property but does not formalize people will inherently tend to serve capital and ownership rather than people.
— Glen Weyl, founder of Radical Markets
Much of what our civilization does collectively depends on how we resolve the identity of the individual. Who is the person seeking a job, buying real estate, or casting a vote? Who is receiving the protection or the judgment of the law? And how do we safeguard this person from powerful interests that might otherwise exploit their vulnerabilities?
The solutions that we have developed to answer these questions, namely state-issued forms of identification, do not entirely address the problem. They are readily given to fraud and abuse, especially where the Corruption Perceptions Index is high. These drawbacks are further compounded on the Internet. There is no governing body to issue identification, so people can profess to be anyone. So can AI, and with increasing persuasiveness. In one recent study, less than half of social media users in the United States report that they are “somewhat confident” that they can distinguish a living human from artificial intelligence. Another study estimates that social media bots are more than twice as influential as human beings in shaping opinions on political figures. Add to this that the form of AI that most commonly resides on social media is relatively unsophisticated. Eventually we will confront more responsive AI that can mimic the subtleties of human interactions and carry on a convincing unscripted conversation.
The privacy issue is even more intractable, and we are just beginning to comprehend its scope. Recent visibility into the policies and practices of multinational companies like Amazon and Facebook has resulted in a degree of public pressure, but the response of the corporate owners of our private data has been unhelpful, if largely predictable. The solution, in their view, is not to return data ownership to the individual but to increase the level of access and oversight of official regulatory bodies. Even in countries with a low perception of corruption, regulators are often political appointments who are difficult to hold to direct account. But in countries with a high perception of corruption, combined with universally criticized repressive governments, what little freedom of speech that social media has afforded is now working to limit that very freedom, an outcome that is beneficial only to demagogues and owners of social media companies. As observers like Alex Gladsetin have reported, in places with longstanding structural human rights abuses, we are already witnessing the vast accumulation of private data by the governing regime, working in tandem with social media providers, to determine who is a potential political danger: in essence, to separate the good citizen from the bad, where “good” carries the meaning of “compliant.”
It has become abundantly clear that our inability to secure our digital identity is not just an online inconvenience. It is actively undoing the promise of human development that the Internet at its best represents. As alternate economies like the blockchain have emerged, with the hope for freer, less corrupt exchanges, participants have found ways to exploit these systems and bend them toward the sorts of outcomes that have decreased confidence and efficiencies in traditional markets: concentration of wealth, unequal access, the stifling of competition, and so on. Likewise, fledgling experiments in direct digital democracy must contend with the reality of a political discourse that has been infected by propaganda-spreading AI and other forms of cyber-sabotage, some of it perpetuated by state-level actors operating in their official capacity, where it is not crushed outright by the loss of private data.
Characteristics of a secure digital identity
To support equality in an e-democracy, a new notion of digital identity must be devised that is truthful, unique, persistent, and owned by the person it represents.
— Ehud Shapiro, Professor of Computer Science and Biology, Weizmann Institute of Science
Ascertaining the ways in which digital identity is insecure helps us to anticipate what is necessary for the solution. In brief, a secure digital identity must possess the following characteristics:
- Uniqueness. A secure digital identity must correspond to one and only one human person.
- Privacy preservation. A secure digital identity must protect the privacy of the human person to whom it refers. The person should remain anonymous in the network; there should be no need to disclose any personally identifiable information of any individual who participates.
- Authenticity. A secure digital identity must be difficult to fake or steal, either by another human or by AI.
- Decentralized online verifiability. There must be a reliable online method to verify a secure digital identity, and this method must not involve a trusted third party or centralized authority, whether public or private. A peer-to-peer verification network is ideal.
This list provides a framework for assessing existing approaches and, crucially, understanding where they are falling short.
Considering solutions
Accurate identification and authentication is still the great unsolved problem of the Internet; we still can’t tell the good guys from the bad…. Our failure to do so is the primary reason why our global economy loses trillions of dollars annually to fraud-related costs.
— Timothy Ruff, CEO and co-founder of Evernym
There is no shortage of solutions for securing digital identity. Some are already in practice, while others remain in the proposal stage. Let us briefly consider the most widespread of these and indicate why they do not adequately address the problem.
Social login
The individual’s identity can be verified by an existing login — most often a social media account such as Facebook, Twitter, or Google — by way of a third-party identity provider, or IDP. OpenID implements this model.
Social login offers a level of convenience to those who are already on the Internet, and it is relatively easy to support, but it does not reflect any of the four characteristics of a secure digital identity. There is no guarantee of uniqueness, privacy, or authenticity among its users, and it relies on the mediation of a trusted third party. It also compounds the problem in the developing world where social media and the Internet are, for all practical purposes, synonymous.
Government ID
The prevailing governing authority can verify the individual’s identity. This is the approach of the Federal Public Key Infrastructure network, or FPKI, in the United States, and similar networks in other countries. It is also indirectly used in Know Your Customer procedures, or KYC, where a private company accepts official documents such as a passport or a driver’s license as proof of identification.
Setting aside that these documents are already widely counterfeited, there is no agreed-upon international standard for implementation on a global scale. Moreover, in places where official corruption is high, the government might not be a reliable third party. And more than 1 billion people around the world have no state-issued identification of any kind.
Biometrics
The unique characteristics of an individual’s body, such as their facial features, fingerprint, voice print, or eye structure, can verify their identity. Apple’s iPhone and other handheld devices make use of biometric authentication.
Unfortunately, this biometric data is hackable and therefore a risk to privacy, and the mediation of the verification software itself introduces a reliance on a trusted third party. Also, adversarial networks are capable of generating credible biometric data, which raises questions of authenticity and uniqueness.
Device-based identity
The possession of a device associated with an individual can be a verification of that individual’s identity, especially of the device is secured by biometrics or some other generally accepted method. If that device appears on the Internet, automatic systems can infer that the individual operating the device is authentic.
But the individual could unlock their device and give it to someone else to use. Or the device could be stolen and subsequently cracked or compromised through malware or a virus. It is also possible to emulate a device virtually on a server. Further, this solution assumes that every individual will be able to own and maintain an Internet device, which could exacerbate material constraints and the distribution of technology in developing countries.
Self-sovereign identity
The self-sovereign identity model, or SSI, has different meanings in different contexts, with varying and sometimes contradictory implications. But in the most widely understood sense, as articulated by Christopher Allen in “The Path to Self-Sovereign Identity,” SSI removes a human arbiter from the verification of identity and replaces it with “an open and transparent algorithm run in a decentralized manner.”
Despite its ideals of openness and transparency, the algorithm itself assumes the function of a trusted authority, which introduces a node in the verification process that can be exploited to limit legitimate access. And many implementations of SSI fall short of the project’s larger goals by relying on tests of identity that are easy to fake and forms of credentials that are easy to counterfeit.
A new way forward
The ideal goal of an identity system is to make it relatively easy for individuals to get one identity, but relatively difficult to get many identities.
— Vitalik Buterin, founder of Ethereum
There have been discussions about whether it is possible to use a reverse Turing test to formalize identity online. This is a simple checkpoint that is relatively easy for a human to pass and relatively difficult for artificial intelligence. But AI is getting better at solving certain kinds of these tests, because they rely on a kind of perception for which it can be trained. And many of these tests are themselves generated by machines. Can we create a test that verifies a participant’s humanity and uniqueness without appealing to a trusted authority? We believe that we can, and this is how.
Verifying the participant’s humanity
To verify the participant’s humanity, the test must be AI-hard, or relatively easy for a human to pass and relatively difficult for artificial intelligence.
We have developed a test that belongs to the class of AI-hard problems, is created by humans, and allows for unpredictability and a theoretically unlimited set of applications: the flip challenge. We envision administering it on a global proof-of-person blockchain.
Verifying the participant’s uniqueness
To verify the participant’s uniqueness, the test must take place at a synchronized time for everyone on the worldwide network. The democratic ideal of one person, one vote on election day becomes one person, one test at validation time.
The user’s identification persists for as long as the current validation period lasts. After the validation expires, participants revalidate with a new synchronized test.
Avoiding the need for a trusted third party
The results of the test must be collectively self-validated. In other words, the network must validate its own users. This can be done statistically if we know how long it takes to complete a flip and the accuracy rate of its completion.
Unique cryptoidentity as the next step to freedom
Determining who has the right to participate and who doesn’t cannot be an afterthought of democracy: It is its elemental task.
A worldwide proof-of-person network is the first step toward implementing a technological infrastructure that serves the interests of people, not the interests of capital. It will vastly reduce the influence of bad actors and serve as a counterweight to repressive regimes; limit the spread of malicious artificial intelligence in human networks; resist the tendency of money and power to accumulate and concentrate; and open the doors to boundless possibilities, including democratic voting, fair value distribution, free speech platforms, unmediated communication tools, authentication without loss of privacy, and self-monetization: in short, the Internet that was promised.
Just as Bitcoin represents a chance to decouple the world economy from the economic hegemony of entrenched interests, expand access to markets and resources, and promote outcomes that are fairer, freer, and more equal than any that we have yet seen in history, Idena represents the same for identity. We envision a project to restore the cryptoidentity of the individual to the private human person to whom it belongs.
This is our idea at a high level. It is not a technical specification, nor is it the final word on the subject. We intend to elaborate our plan.
In the weeks and months to come, we will consider individual facets of the solution more closely. Follow our blog to get updates.
***
Idena is a novel way to formalize people on the blockchain. It does not collect or store personally identifiable information. Idena proves the humanness and uniqueness of its participants by running an AI-hard Turing test at the same time for everyone around the globe.
The Idena blockchain is driven by proof-of-person consensus: Every node is linked to a cryptoidentity — one single person with equal voting power.
