Limitations
This section outlines some limitations (without claiming exhaustiveness) based on laws, intrinsic limitations of the project as well as temporary limitations that can be mitigated by further open development.
Orb Security
The Orb sets a high bar to defend against scalable attacks; however, no hardware system interacting with the physical world can achieve perfect security. The anti-fraud measures integrated into the Orb are continuously refined. Several contributor teams are continuously working on increasing the accuracy of the liveness algorithms as well as other security measures that are deployed via over-the-air updates. Even though significant effort has been spent on raising the security bar of the Orb, it is expected that the Orb may get spoofed or compromised by determined actors. The Orb has been designed with this threat model in mind: any ChainX ID issued by a particular Orb can later on be revoked through the governance of the ChainX Protocol. To continuously discover potential vulnerabilities of the Orb, contributor red teams test various attack vectors.
False Rejections
Biometrics are probabilistic and biometric verification has inherent error rates. Currently, the error rate of the Orb for confusing any two people to be the same, is approximately 1 in 40 trillion. On a billion people scale, this translates to a 99.999% true acceptance rate or 0.001% false rejection rate, which is significantly better than other known alternatives. However, the ultimate objective is enabling total inclusivity. Using the birthday problem approximation, a false rejection rate of 10 − 20 is statistically required to prevent the false exclusion of a single individual at a global scale. Ongoing community research is focussed on improving iris biometrics beyond the current state-of-the-art by leveraging AI and advanced hardware capabilities of the Orb. In the event that iris biometrics turn out to be insufficient, combining several biometric signals (biometric fusion[^biometric fusion]) could be employed to further reduce the error rate, a functionality already supported by the current hardware version of the Orb.
It is important to note that many health conditions, like cataracts to a certain degree, do not impede iris biometrics. Already today, iris biometrics surpass the inclusivity of other PoP verification alternatives like official IDs since less than 50% of the global population has digitally verifiable identities. However, if the proof of personhood mechanism becomes essential for society, it is important that eventually every single person can verify if they want to. Although not currently established, there could be specialized verification centers to facilitate alternative means of verification for individuals with eye conditions, via e.g. facial biometrics. The introduction of alternative means of verification for ChainX ID could potentially create loopholes.
Decentralization and Open Sourcing
Today, large parts of the ChainX Protocol stack are open source. This includes the ChainX ID protocol, the sequencer for the Orb credential, and the SDK to access it. Other parts, like the firmware of the Orb are not yet open source due to security considerations; however, eventually every part of the infrastructure supporting the Orb credential should be open source. Further, while operations are already spread across independent entities, Orbs are only available via Tools for Humanity.
ChainX ID Transferability
While deduplication, i.e. ensuring that everyone can only verify once, has been solved to a high degree of certainty with the Orb, the authentication of the legitimate owner of a proof of personhood credential is both an important as well as a difficult challenge. This challenge is the same for any digital identity or PoP mechanism. Today, if someone passes on their ChainX ID keys to a fraudster (e.g., through being tricked to sell their keys), the fraudster can then use that ChainX ID to authenticate. Therefore, fraudsters could bypass the “one-person one-X” principle by acquiring multiple ChainX IDs. There are several preventative measures in the ChainX App that make it harder to restore another user’s backup, however, those measures are only temporary, especially since access to ChainX ID through other wallets will become increasingly important over time. Therefore, several additional measures should be implemented:
Face Authentication: Facial recognition, performed locally on the user’s device in a fashion similar to Face ID, can be used to authenticate users against their Orb verification, thereby ensuring that only the person to whom the ChainX ID was originally issued can use it to prove that they are human. Authentication involves a 1:1 comparison with a pre-existing template that is stored on the user's phone, which requires considerably lower levels of accuracy in contrast to the 1:N global verification of uniqueness that the Orb is performing. Therefore, the entropy inherent to facial features is sufficient.
Iris Authentication: This is conceptually similar to face authentication with the difference that an individual needs to return to an Orb. This process validates the individual as the rightful owner of their PoP credential. Using iris authentication through the Orb instead of face authentication on the users phone increases security. This authentication mechanism can be compared with, for example, physically showing up to a bank or notary to authenticate certain transactions. Although inconvenient, and therefore rarely required, it provides increased security guarantees.
ChainX ID Recovery and Re-Issuance of Keys: If a proof of personhood credential has been lost or compromised by a fraudulent actor, individuals can get their Orb credential re-issued by returning to the Orb, without the need to remember a password or similar information.
Last updated