Nicolai Baldin, Co-founder and CEO, Synthesized.
The so-called FinTech revolution has long been in progress, with a number of challenger banks emerging as its patrons. It reflected a growing sense of unrest among the general population for a more transparent, seamless and cost-efficient means of banking. With stronger regulations in place, consumers feel within their rights to demand better safeguarding of personal data, and clearer limits on what one bank can and cannot do with their records. Yet, until the pandemic, transformation had taken a piecemeal approach compared to what we have witnessed since.
Acting as the incendiary of change, the sweeping impact of Covid-19 has compelled millions of individuals across the globe to seek loans, conduct transactions and manage their finances from home. Overnight, banks and other financial institutions have had to accommodate for this new reality with a plethora of digital solutions; not only to maintain a competitive edge but to stay in the running altogether.
Of course, it is not enough to throw a number of solutions at customers and see what sticks. Developing such technologies is a heavy investment and financial organisations need to meet the ever evolving needs of today’s consumers. Better still, institutions should be anticipating wants and needs before customers even realise the desire, themselves. In order to do so, these institutions need data, and more importantly, the ability to draw valuable insights from it. Data is fundamental to the success of all businesses but it is only as useful as the information drawn from it. At its core that provides access and the ability to work collaboratively on the data with trusted partners.
Synthesized recently conducted a global survey and found that the ability to share and collaborate internally as well as externally on sensitive data, in other words customer data, ranked 8.3 out of 10 for organisations, with 10 indicating high importance. Yet, many also have to contend with three primary challenges.
First, maintaining data privacy compliance (ranked 7.44); followed by data security, particularly with the rise in cyber threats (ranked 7.12); and third, clear data governance when sharing data with external parties (ranked 6.64).
With so much at isk, such as falling in breach of regulations, potential fines and reputational damage if data were to leak, be breached, or stolen, some organisations may be reluctant to let employees easily access and share data, introducing a significant opportunity cost. So, what’s to be done? How can organisations facilitate greater collaboration whilst ensuring the privacy and safety of the data they hold? The simple answer to this is artificial intelligence – leverage AI to automatically create at any scale high-quality data products that are GDPR compliant, safe to share, while maintaining the statistical properties of original (customer) data.
Using the power of data synthesization technology, the AI engine takes a set of data, understands its characteristics, behaviours and general look, and produces a precise replica, with all personal attributes removed. It creates simulated data that is highly representative of the initial dataset and in addition, ensures the new data set is statistically clean, complete and accurate.
Through deep learning, some tools may even have the capacity to identify and mitigate biases in real-time, augmenting the data to ensure a fairer representation. In addition, such tools enable data practitioners to rigorously test any scenarios and improve test coverage to a much broader range than original data could ever provide.
Unlike other techniques such as data masking or obfuscation, the simulated intelligent data cannot be easily reverse-engineered. Naturally, in doing so, organisations essentially remove the object of risk. By removing Personally Identifiable Information (PII), organisations can further sidestep the vast majority of regulations that would have otherwise impeded the use of, as well as access to, data. With financial institutions harbouring a wealth of highly sensitive information this is a great advantage.
Taking this a step further, organisations may also want to consider the use of data clean rooms, or isolated and secure environments in which companies can safely share their data with remote internal teams or even third-parties, whilst maintaining control over data access and movement. In this way, clean rooms facilitate collaboration internally as well as externally by providing a safe environment to brainstorm ideas and innovate, while synthetic data itself is the tool which allows organisations to easily balance better collaboration, data security and privacy.
