Here you can find a short video interview with David Roberts, Technical Sales at Privitar, in which he talks about some of the challenges of adopting a Data as a Service (DaaS) approach, and how best to overcome them.
[0:08] Why do enterprises adopt Data as a Service?
So, I think the reason that all enterprises are adopting data as a service (DaaS) platforms is fundamentally just to get more insight out of their data. They can pull that data together from all the different silos around the organization, enrich it and then extract more insightful data-driven decisions from that data. Secondly, it's just to get control over how that data is used and provisioned throughout the organization as well as to third parties and enable them to audit and control the use of access of the data.
[0:40] What are the challenges of adopting Data as a Service?
One of the biggest challenges is how do you protect the privacy of individuals represented within these data sets once they're all brought together, as well as maintaining the usefulness and utility of that data to the teams that want to extract insight from it (e.g. analytics teams, data science, marketing and so on). I think secondly is how do you manage that at scale: how do you cater for the demand from all the teams that want access to that to that data, as well as the volume and rate of information that's flowing into that data set. Thirdly, simply by bringing together data into one place, not only you do increase the value of that data, but you also increase the risk and implications of what may happen if there's a data breach, either maliciously or accidental from internal users.
[1:35] How does Privitar help with these challenges?
Given those challenges, the way Privitar helps our customers addressing the balance between preserving usefulness and utility from the data as well as protecting individuals' identity, is to provide a breadth of techniques: ranging from masking, perturbation, clipping, as well as more advanced techniques like generalization and differential privacy. Then, to help customers manage their data governance at scale we provide a set of tools to allow them to manage privacy policies holistically across the whole data set, as well as embedding watermarks within those data sets: in this way, at a later date, they can retrieve information like who the data was given to, what it contained, when it was created and so on.
And then, to address the risk of internal breaches and minimize the attack surface for data sets, we are also able to sandbox data sets using what we call Protected Data Domains (PDDs), and this allows individual sets of data to be provisioned to teams, so that they cannot be joined and individuals within those data sets re-identified.
[2:48] How is Privitar's solution different?
So, I think one area where Privitar is different, is that we actively work with a number of the leading academic research organizations around the world in the area of privacy engineering, and we can bring that leading edge state-of-the-art thought and research into the enterprise world, where our customers can instead of treating privacy as a burden that has to be complied with can embed it into their processes and systems, and really treat privacy as a competitive advantage.