Leveraging AI to improve review speed without sacrificing quality
In civil litigation in the U.S., defendants and plaintiffs must disclose evidence to each other prior to the hearing in order to sort out issues, and electronic data and documents that can serve as evidence must be submitted by the due date in accordance with appropriate procedures. In particular, the process of handling electronic data is called e-discovery, and if a Japanese company does business in the U.S., all of its electronic data at its headquarters or data center in Japan may be subject to disclosure as evidence. Once involved in a lawsuit, the necessary information must be quickly and appropriately sorted from the vast amount of data in order to avoid creating a litigious disadvantage.
KIBIT Automator is an AI-based tool developed based on the opinions and requests from various investigative agencies and the knowledge and know-how accumulated through FRONTEO's abundant experience in forensic investigation. KIBIT Automator is an AI-powered tool developed independently based on our accumulated knowledge and expertise in e-discovery investigations. The most expensive part of the e-discovery process is the document review process, which is estimated to account for about 70% of all discovery costs. The challenge for companies in discovery is how to streamline this review process and reduce costs.
Traditionally, clients and law firms have raised concerns about document review, including cost, time, volume, and the speed and skill of reviewers. In addition, because the fee model used to be based on a man-month basis, some clients were forced to compromise on quality for more difficult reviews in order to keep costs down. KIBIT Automator" reduces the time required for reviews by implementing the functions described below, and enables document reviews with quality that is not affected by reviewer skills or fatigue caused by long hours of work, thereby improving the efficiency of review operations without sacrificing quality.
Main Functions
Assisted Learning: using a cutoff assistance function to determine "documents that do not need to be read"
greatly reduces the number of documents that must be read by a human being.
Normally, document review is handled by attorneys or people working under the direction of attorneys, who review documents one by one to determine what percentage of the documents collected for study need to be reviewed to find a certain percentage of "relevant" documents, using the "relevant" Assisted Learning function. The result is a "relevant" document that can be used by the attorney in charge of the case or by the client to determine whether the document is "relevant. As a result, the attorney in charge of the case or the person in charge of the company can exclude documents from the list of documents collected for investigation as "documents that need not be read" or give priority to documents that should be read, thereby shortening the time required for investigation.
Highlight Sentence: Highlight the part that needs confirmation
At the time a task is sent to a reviewer, the highlighted sentences are passed on to the reviewer, reducing the amount of text that the reviewer must actually read.