It is now several months since the GDPR came into effect. Worldwide, organizations have now been able to get a fair sense of how data protection works under the new policy. While the new regulation has been effective in staving off potential data misuse by at least among regulated businesses in the EU and worldwide, the loss of data due to security breaches and from individuals and countries not under the purview of the law continues unabated.
As organizations double down on their consumption of data and analytics in their decision making, the threat of security breaches and the regulations around data protection are two areas that are going to routinely come under the scanner.
Take the example of the ransom-ware attack that hit the Bristol airport a few months back that blacked out all the displays at the terminus. It is not far fetched to assume the likelihood of such attacks on other computer systems within the airport or elsewhere. This could expose critical financial data of millions of passengers, and also expose cities to cyber-based terror attacks.
So how exactly do organizations handle data analytics in the age of data protection and cyber threats?
One of the most effective ways to do this is through consolidation. According to an IDG report, unstructured data is growing at the rate of 62% annually and it is estimated that by 2022, nearly 93% of all data is unstructured. Unstructured data is not only more voluminous, but is also more vulnerable to security threats. This is because securing data costs money and this expense rises with the volume of data to protect.
Many organizations today invest in what is known as analytics stacking - this is the transformation of raw and unstructured data into its structured and consolidated form that makes it valuable for business intelligence. Such structured pieces of data is more valuable than raw data. More importantly, it occupies a stunningly lower volume and this makes protection cheaper and effective.
However, it is not always possible to proactively transform all unstructured data into structured data for analytic purposes. This is especially true if an organization deals with a wide variety of data sources that have uniquely different purposes.
Such instances call for the use of forensic data analytics (FDA). In this case, your analytical tools query each of your incoming pieces of data to identify patterns that may be used to identify its utility. In short, such tools make it unnecessary for your organization to provide sampling data and instead deduce patterns based on previous inputs.
While protecting your business data from cyber-criminals is top priority, some organizations believe that data protection laws tend to hamper innovation. A report published by the Center of Data Innovation finds GDPR regulations governing AI to be slowing down research and innovation. The report specifically points to regulations surrounding the need for companies to manually review significant algorithmic decisions that could raise the cost of AI while the right to explanation could bring down accuracy.
In effect, data privacy regulations could potentially impact the trajectory that modern technologies like AI and machine learning could take. Some experts also warn that the need for machine learning algorithms to explain their output (as stipulated by GDPR) could make deep learning illegal. However, others point out that these strict regulations have paved the way for a more regulated nurturing and growth of AI/ML. In the past year, EU countries like France, Germany, Britain and Denmark have unveiled their own draft policies that would enable furthering research in the industry while ensuring that the technology is used ethically by corporates and governments alike.
Regardless of the massive strides we have taken in the areas of data analytics and AI/ML, the truth is that the industry itself is still at its infancy. The potential use-cases for data analytics, AI and ML is massive and is only going to grow in the next few decades. Regulations and the threat from cyber-criminals will help ensure that organizations tread a fine line and keep customer data secure while growing their own capacities with these technologies.