Check out all the on-demand sessions from the Intelligent Security Summit here.
Data protection is challenging for many businesses because the United States does not currently have a national privacy law — like the EU’s GDPR — that explicitly outlines the means for protection. Lacking a federal referendum, several states have signed comprehensive data privacy measures into law. The California Privacy Rights Act (CPRA) will replace the state’s current privacy law and take effect on January 1, 2023, as will the Virginia Consumer Data Protection Act (VCDPA). The Colorado Privacy Act (CPA) will commence on July 1, 2023, while the Utah Consumer Privacy Act (UCPA) begins on December 31, 2023.
For companies doing business in California, Virginia, Colorado and Utah* — or any combination of the four — it is essential for them to understand the nuances of the laws to ensure they are meeting protection requirements and maintaining compliance at all times.
Understanding how data privacy laws intersect is challenging
While the spirit of these four states’ data privacy laws is to achieve more comprehensive data protection, there are important nuances organizations must sort out to ensure compliance. For example, Utah does not require covered businesses to conduct data protection assessments — audits of how a company protects data to determine potential risks. Virginia, California and Colorado do require assessments but vary in the reasons why a company may have to take one.
Virginia requires companies to undergo data protection assessments to process personal data for advertising, sale of personal data, processing sensitive data, or processing consumer profiling purposes. The VCDPA also mandates an assessment for “processing activities involving personal data that present a heightened risk of harm to consumers.” However, the law does not explicitly define what it considers to be “heightened risk.” Colorado requires assessments like Virginia, but excludes profiling as a reason for such assessments.
Intelligent Security Summit On-Demand
Learn the critical role of AI & ML in cybersecurity and industry specific case studies. Watch on-demand sessions today.
Similarly, the CPRA requires annual data protection assessments for activities that pose significant risks to consumers but does not outline what constitutes “significant” risks. That definition will be made through a rule-making process via the California Privacy Protection Agency (CPPA).
The state laws also have variances related to whether a data protection assessment required by one law is transferable to another. For example, let’s say an organization must adhere to VCDPA and another state privacy law. If that business undergoes a data protection assessment with similar or more stringent requirements, VCDPA will recognize the other assessment as satisfying their requirements. However, businesses under the CPA do not have that luxury — Colorado only recognizes its assessment requirements to meet compliance.
Another area where the laws differ is how each defines sensitive data. The CPRA’s definition is extensive and includes a subset called sensitive personal information. The VCDPA and CPA are more similar and have fewer sensitive data categories. However, their approaches to sensitive data are not identical. For example, the CPA views information about a consumer’s sex life and mental and physical health conditions as sensitive data, whereas VCDPA does not. Conversely, Virginia considers a consumer’s geolocation information sensitive data, while Colorado does not. A business that must adhere to each law will have to determine what data is deemed sensitive for each state in which it operates.
There are also variances in the four privacy laws related to rule-making. In Colorado and Utah, rule-making will be at the discretion of the attorney general. Virginia will form a board consisting of government representatives, business people and privacy experts to address rule-making. California will engage in rule-making through the CPPA.
The aforementioned represents just some variances between the four laws — there are more. What is clear is that maintaining compliance with multiple laws will be challenging for most organizations, but there are clear measures companies can take to cut through the complexity.
Overcoming ambiguity through proactive data privacy protection
Without a national privacy law to serve as a baseline for data protection expectations, it is important for organizations that operate under multiple state privacy laws to take the appropriate steps to ensure data is secure regardless of regulations. Here are five tips.
Partner with compliance and legal experts
It is critical to have someone on staff or to serve as a consultant who understands privacy laws and can guide an organization through the process. In addition to compliance expertise, legal advice will be a must to help navigate every aspect of the new policies.
Identify data risk
From the moment a business creates or receives data from an outside source, organizations must first determine its risk based on the level of sensitivity. The initial determination lays the groundwork for the means by which organizations protect data. As a general rule, the more sensitive the data, the more stringent the protection methods should be.
Create policies for data protection
Every organization should have clear and enforceable policies for how it will protect data. Those policies are based on various factors, including regulatory mandates. However, policies should attempt to protect data in a manner that exceeds the compliance mandates, as regulations are often amended to require more stringent protection. Doing so allows organizations to maintain compliance and stay ahead of the curve.
Integrate data protection in the analytics pipeline
The data analytics pipeline is being built in the cloud, where raw data is converted into usable, highly valuable business insight. For compliance reasons, businesses must protect data throughout its lifecycle in the pipeline. This implies that sensitive data must be transformed as soon as it enters the pipeline and then stays in a de-identified state. The data analytics pipeline is a target for cybercriminals because, traditionally, data can only be processed as it moves downstream in the clear. Employing best-in-class protection methods — such as data masking, tokenization and encryption — is integral to securing data as it enters the pipeline and preventing exposure that can put organizations out of compliance or worse.
Implement privacy-enhanced computation
Organizations extract tremendous value from data by processing it with state-of-the-art analytics tools readily available in the cloud. Privacy-enhancing computation (PEC) techniques allow that data to be processed without exposing it in the clear. This enables advanced-use cases where data processors can pool data from multiple sources to gain deeper insights.
The adage, “An ounce of prevention is worth a pound of cure,” is undoubtedly valid for data protection — especially when protection is tied to maintaining compliance. For organizations that fall under any upcoming data privacy laws, the key to compliance is creating an environment where data protection methods are more stringent than required by law. Any work done now to manage the complexity of compliance will only benefit an organization in the long term.
*Since writing this article, Connecticut became the fifth state to pass a consumer data privacy law.
Ameesh Divatia is the cofounder and CEO of Baffle
Welcome to the VentureBeat community!
DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own!