Enhancing Identity Verification Through Artificial Intelligence

October 2020
As the need for and scale of identity verification rapidly expands, it is becoming increasingly apparent that traditional options that we have relied on for verification and validation are not up to the task. Instead, we find ourselves turning to technology for our solutions. There are a number of ways in which technology can be utilised to improve the identity verification process, but one of the key areas that has been implemented is through Artificial Intelligence (AI).

As with any new technology that comes into use, there have to be compelling reasons and benefits for it to gain traction and find use, and AI is no exception. One of the main factors in AI’s favour is the sheer volume of data it can get through. Compared to traditional human verifiers, AI has the advantage of being able to work more or less 24 hours a day, 7 days a week, 365 days a year. Even discounting the fact that a programme doesn’t need sleep or rest or breaks, the processing ability of a suitable AI solution is also able to analyse and sort data at a much faster rate than any human could ever hope to. In addition to the speed at which it can process vast amounts of data, AI solutions are shown to be much more accurate than human verifiers, being able to immediately discern the validity of documentation and matching ownership. Linked to this, the benefits of AI go beyond the purely technical and into the emotional. When documents are submitted for verification, the data on them is likely to include personal information. Because of the sensitive nature of the data being assessed, people may feel uncomfortable with another human being seeing and processing that information as, whether malicious or not, there will be some level of judgement, or if not judgement then at least recognition. When an AI programme verifies identity, there is a complete lack of judgement - there is no emotional intelligence behind the AI and therefore an individual’s data remains secure and personal to them.
Beyond the immediate improvements that an AI solution holds over traditional options, there is an important benefit to note in that the people who were previously tied up in performing laborious verification services are then freed up to work on more important tasks. Following on from this, there is also a benefit to removing the human element from the verification process in that any burden of legal risk that would fall on individuals performing identity verification is no longer an issue.
Although the benefits of AI identity verification solutions are manifold, it is important to note that there are still downsides. The most obvious potential issue that could arise from an AI solution is the coding. A programme that is improperly coded and full of bugs will obviously not be up to scratch and will hamper the verification process rather than improve it. A company looking to implement an AI solution therefore needs to ensure that the product they choose to use is of high quality and that they can trust the technology to handle the workload. The second thing that should be noted is the inability for an AI solution to deal with something that isn’t covered by its code. This can be an issue in cases where documentation has not been standardised in layout, involves different languages and alphabets, or has different data fields to what the program is written to accommodate. In these situations, the solution will require adapting or other solutions to work in tandem with it before it can process the data. In addition to this, no solution is ever perfect and there may be cases where an AI solution is unable to decide if an identity passes the verification process. In these instances, a human operator will need to take over and decide the outcome; however, in most solutions this is expected to involve only a very small proportion of cases and the workload placed on human workers will still be drastically reduced. Due to these limitations of AI, it is recommended by most that any identity verification system still involves humans at some level to deal with the few areas in which an AI programme can’t determine an outcome.   
AI is a significant technological development, but there are numerous other technologies that have the potential to intersect with it. Areas such as Machine Learning can be used to improve AI’s capabilities without requiring a high level of human oversight. Using AI allows Big Data to be accessed in a realistic time frame, enabling more accurate decision making and further-reaching analysis. Cutting edge research into cryptography may lead to a future where personal data can remain completely private whilst still enabling identity verification with 100% confidence in the accuracy and validity of the results. And of course, with AI solutions and developments that stem from them, the world of blockchain and smart contracts can interface with human elements where individuals need to be represented legally in a digital ecosystem. 
At Blockpass we utilise AI verification solutions as part of our verification and onboarding services. The KYC process is a prime example of a verification requirement that is becoming more commonplace and urgent as the way we interact and conduct business evolves. By implementing AI and a number of other technological innovations, and with our research into new cryptographic areas in conjunction with Edinburgh Napier University at the Blockpass Identity Lab, we are working to make verification and onboarding as quick, efficient and painless as possible.
The Blockpass platform is fully automated and hosted in the cloud, with no integration or setup fee. Businesses can sign up to the KYC Connect® console in a matter of minutes, test out the service, and start conducting identity documents verification, KYC and AML checks. Sign up for FREE at