In the early days of discovery, finding—and then protecting—sensitive personal information within evidence for litigation or an investigation was a manual process involving hands-on reviewers with black markers and a keen eye for detail.
In the current days of e-discovery, finding and protecting sensitive personal information is, for many teams, still a manual process, but instead of black markers, it involves hands-on reviewers with RegEx search formulas, redaction software, and a keen eye for detail.
But now, astronomical data volumes, tight deadlines, and the evolving landscape of regulations mean that protecting the privacy of personal data is a more high-stakes activity than ever before.
Fortunately, RelativityOne users have two robust options to simplify and accelerate this process: Redact and Personal Information Detect. But what are the differences between these two tools, and how do you know when to use one over the other?
That’s our topic for today. Keep reading to learn more about each of these options, and when one might be more appropriate than the other.
Redact in RelativityOne
Redact is an advanced redaction solution we released back in 2021.
Using Redact, e-discovery teams can protect personal information in their data sets with automated redactions on several file types, including images, spreadsheets, and PDFs. To apply these redactions, users follow three steps:
- Define project rules. These are created based on known formulas for personal information, such as Social Security numbers; addresses; phone numbers; and so on. Regular expressions are used to search for and identify instances of these known patterns in personally identifiable information.
- Begin running automatic redactions. Redact will then search through your documents and redact (or highlight) data according to the set of rules for the project.
- Perform QC on Redact’s results. Once the project has finished applying redactions, users will see a project summary with flags for potentially questionable documents or results. Then, the markups can be approved or rejected within the Viewer using mass operations.
This feature is directly integrated into RelativityOne, and available to all users of the platform at no additional cost.
Relativity Personal Information (PI) Detect
New this year, Relativity PI Detect is an AI-powered solution for uncovering and protecting sensitive personal information hidden in a data set.
PI Detect leverages machine learning and natural language processing to help you find this information in a way that goes beyond traditional searching. The product uses more than 120 pre-trained AI detectors—available in multiple languages—to spot personal information even when it doesn’t adhere to traditional patterns.
After selecting the AI detectors to use and running a document set through the product, the personal information is highlighted on the documents and a document report is generated with the full set of identified PII. Additionally, the AI can be re-run throughout a matter and some detectors are able to learn based on context and reviewer changes—making it more effective within the matter.
Once this information is identified, PI Detect uses façade redactions which automatically marks up the document based on the identified personal information.
PI Detect is fully integrated into RelativityOne, and available to add on to your existing subscription at an additional cost.
When is PI Detect Most Useful?
Sometimes, you need a magnifying glass. Other times, you need a microscope. PI Detect is that microscope, offering more robust and powerful capabilities for combing through large data sets for the small but essential details that need to be protected from wandering eyes.
PI Detect is an ideal solution if you:
- Are working on a matter with a high likelihood of containing personal data, like individuals’ health or contact information
- Need to complete thorough redactions on a short timeline
- Believe traditional searching will miss personal information that uses incorrect formatting or misspelled details
Matters involving healthcare, alleged discrimination or harassment, class action, or product liability can span huge data sets, and sometimes, that data is rich with personal information. In cases like these, the added sensitivity of AI-powered PII detection can make a significant difference in finding and redacting personal information with greater accuracy and in less time.
How to Get Started Effectively Safeguarding Personal Information
Corporations and law firms are entrusted with the protection of immense amounts of personally identifiable information pertaining to their clients, consumers, and employees—and increasing demand from customers and regulators to ensure PII remains safe adds additional pressure. Protecting that PII is essential not just from an ethical perspective, but from a legal one.
e-Discovery can be a particularly risky activity when it comes to the protection of personal data; after all, documents must be collected, ingested, analyzed, reviewed, produced, and presented by various individuals at various stages of litigation. Automating more of these touchpoints can help minimize the risk of human error, and uncover more insights in less time.
If you have questions about whether Redact or PI Detect is a better fit for your next project, or want to take a look at either product, please reach out. We’re here to help.
Hannah joined Relativity in 2019 and is the Senior product manager. Hannah is focused on building AI-powered products to support Privacy use cases like Personal Information Detection and Data Breach Response. Prior to her work on the Privacy team, Hannah worked with Chris Brown as his Chief of Staff supporting key product and technical initiatives. Her former experience also includes strategy work advising on Cloud and Agile transformation.