How AI and Machine Learning Are Impacting the Litigation Landscape

avatar
by Lindsey Francy Mar 21, 2023 News
How AI and Machine Learning Are Impacting the Litigation Landscape

Artificial intelligence has been present in our everyday activities for a long time. The power of artificial intelligence was made accessible to anyone with a web browser, thanks to the public reveal of the project. Artificial intelligence and machine learning are being used more and more in litigation, especially when it comes to expert witness preparation.

Leading-edge analytical tools and data science techniques are required for the support of expert witnesses. The idea of technology being able to think and make decisions, accomplish tasks more quickly and with better results than humans, conjures up images of a "Jetsons-like" world run byrobots. The "futuristic" ideas about the impact of artificial intelligence were not that far off from a reality. Artificial intelligence has evolved into machine learning where computers are programmed to accurately predict outcomes by learning from large data sets.

Law firms and economic and financial consulting firms have known about the power of artificial intelligence and machine learning for a long time. Artificial intelligence is suited to support, qualify, and substantiate expert work in litigation matters, which used to be done manually. Artificial intelligence and machine learning have been used in expert testimony by both sides.

Humans are partly to blame for the increased use of artificial intelligence and machine learning in expert work as we produce ever-growing volumes of user-generated content. Consumer reviews and social media posts are becoming more relevant in litigation. One approach that can be used is to identify a more manageable subset of the data. This is limiting as it often produces results that are irrelevant to the case whileomitting relevant results. The entire text can be considered using context and syntax to identify the linguistic elements that most accurately indicate relevance.

If you want to see this approach in action, consider litigation involving alleged marketing misrepresentations or defamation. The most robust analyses make them ideal for outsourcing to the non controversial training data and impartial models that are hallmarks of state-of-the-art artificial intelligence and machine learning approaches.

Artificial intelligence and machine learning have proven to be useful in consumer fraud and product liability matters. Humans have the ability to adapt a solution to other uses. These novel uses can be found here.

Publicly available sentiment models perform well on many problems, but fail on tasks with domain-specific linguistic structures. When tasked with measuring the sentiment surrounding an entity in an industry whose discussion features novel or counterintuitive language, failure may arise. Consider a defamation case. In the fitness space, the term "confusion," "resistance," and "to failure" are used to describe a successful workout. Slang terms like "guns" and "shredded" are completely different in the fitness context. Training a domain-specific sentiment model will give a more accurate assessment of the sentiment contained in the statements. The training process would involve gathering hundreds of thousands of user-generated reviews for industry products and then directing a context-aware language model to predict the review score from the text. Through time and around certain critical events, this custom model will be able to track the discussion surrounding theinfluencer.

Assessing marketing influence on social media can be done using artificial intelligence and machine learning. Language models and text similarity metrics can be used to assess if earned media immediately following the company's posts were more like the company's posts than either earned media preceding the posts or picked at random.

A custom object detection model can be trained and applied to a random sample of millions of social media images.

Artificial intelligence and machine learning can be used to quantify the extent and timing of public awareness of a marketing claim. The at-issue topic can be isolated from other related topics. It is possible for an analysis to be more focused on the claim at hand.

Artificial intelligence and machine learning can be used to describe a company's social media presence. A model trained on text and image content from brands that are not related to one another can learn to differentiate between the two. It is possible to quantify whether the brand features are conveyed by the social media content.

Defamation is hard to prove even in the presence of clearly negative statements. It is possible for defendants to claim that the statements were entertainment or satire. By using data and models, experts can objectively measure the degree to which reasonable consumers will interpret the information as fact.

Product liability is a growing area of research. Product liability cases can look at user-generated product reviews to identify at-Issue product features. Rather than assess the review as a whole, aspect-based sentiment analysis focuses on at-issue features only.

A successful class certification challenge will show that the circumstances of the class members were different. The methods can be combined to quantify the heterogeneity of the materials. In a case of marketing misrepresentations, it is possible to train a classifier to distinguish at-Issue marketing content from content not at issue, model the topics targeted throughout multiple distinct marketing campaigns, and summarize images to demonstrate different appeal to different consumers.

Humans have been able to mold resources to serve their needs for hundreds of years. The above examples show it in our small corner of the world. As the availability of user-generated data continues to expand and become more complex, we will continue to see it. Artificial intelligence and machine learning help us find the needle in the haystack. The people who try to find the needle by hand will never be found.

The article was first published by Law.com.

Cornerstone Research's views are not represented by the views expressed.