While Artificial Intelligence (AI) has been used for years in the eDiscovery space, we’ll explore how AI can be further used to automate tasks as well as improve efficiency in eDiscovery workflows. We’ll also explore what’s on the horizon regarding AI use in discovery proceedings.
Although AI isn’t a new concept in the legal industry, there are still some people who haven’t added it to their arsenal of tools to aid in decision making. With the expanding number of problems and data types that machine learning can address, we want to encourage people who aren’t already taking advantage of the power of AI in eDiscovery to learn more about it. The benefits of AI for eDiscovery are numerous, and here is a short list of things you might find interesting. AI can:
- Increase efficiency by automating manual tasks and allowing you to focus on high-value work.
“Finally, more time to plan case strategies”
- Reduce costs by eliminating the need for increased hours reviewing documents or additional staff members during a time crunch.
“Saving our clients time and money? Score!”
- Improve quality by reducing errors in the review process; Often AI will correctly identify relevant documents like a human review while reducing relevant documents marked not relevant (false negative) and can minimize “over marking” borderline relevance or not relevant documents (false positives). In fact, AI will very likely surface important documents that are present in the collection in smaller numbers.
“Minimizing manual review? I’m listening…”
AI Technologies Used for eDiscovery
There are several AI technologies that are commonly used to enhance eDiscovery processes. These include:
- Natural Language Processing (NLP) – This type of AI understands speech and communication as a human would
“Alexa, teach me more about AI…”
- Machine Learning (ML) – finds similar documents based on prior review choices and criteria in the set
Netflix recommends – Partner Track
eDiscovery Use Cases
Here are a few examples of how AI can be used in eDiscovery work:
- Legal research. As the volume of data increases, so does the need to search and analyze it. AI can help lawyers find relevant information faster than people trying to guess which keywords would be relevant by quickly identifying patterns in large datasets. This is especially true with unstructured data such as emails or social media posts, where understanding context is key to finding relevant information.
- Document review. During litigation discovery, documents are often pulled from databases and reviewed by human lawyers (linear review) before being handed over to opposing counsel or released publicly. While many litigators and in house teams are now comfortable utilizing AI, depending on how it is applied and the reliance on its results, AI use may need to be covered in a negotiated, court approved ESI protocol. . With AI’s ability to sift through millions of documents quickly and correctly identify those that are most likely to contain relevant information, this task has become more commonplace and mutually advantageous for the parties involved in litigation proceedings.
But wait, there’s more…
- Contracts analytics
- Matter management
- Knowledge management
- Governance and compliance monitoring
- AI assistants
- Patent searches
- Lar firm invoice review
- Document drafting
And the list goes on.
As we see AI in the legal industry improving processes and speeding access to actionable information, it’s advancing eDiscovery work. Artificial intelligence is amazing, but it’s just that. Artificial. There isn’t anything, yet – and big on the yet – that perfectly imitates the way the human brain works. Since AI is still developing it does carry its own set of challenges and should be applied where it is most effective and can be guided by a human.
Challenges of AI in eDiscovery
The first is data security and privacy issues. AI systems are trained on large amounts of data, which means they need access to sensitive information to function properly. As we have seen from recent events like the Facebook/Cambridge Analytica scandal, this can be a problem when that sensitive information ends up in the wrong hands or is used for nefarious purposes by third parties who have access to it.
Another challenge is lack of human oversight: while an AI system may be able to identify patterns in your documents that you didn’t know existed before (and therefore help you find relevant information, there will always be scenarios where humans need to make judgments about what constitutes relevance or admissibility–and whether something should be kept confidential because it could reveal private details about someone else involved in litigation. Therefore, we see review workflows still requiring human oversight. Additionally, with AI requiring human oversight, comes bias. Humans are innately bias individuals which can affect the consistency and efficiency of a review.
The Future of AI and eDiscovery
The future of AI and eDiscovery is bright. As technology continues to advance, it will bring about new ways of analyzing data that are more efficient and effective than ever before. Companies can expect continued automation, improved performance, and increased efficiency when using AI-powered solutions for their eDiscovery needs.
AI is a powerful tool for eDiscovery and can be used in many places in the process to improve efficiency and quality, but it is not without its flaws. The benefits of AI include the ability to process large amounts of data quickly and efficiently–and even more importantly, in a way that humans cannot do alone. It also allows for greater efficiency in searching through documents because it can identify patterns in text that humans might miss.