The system has been processing data from 70,000 cases, contributing to its increased accuracy
Fletchers, a law firm in the UK leading the development of artificial intelligence (AI) technology for civil case assessment, has announced that its AI-driven claims-sorting system has achieved a 91 percent success rate in identifying unmeritorious claims.
This breakthrough is part of the firm’s ongoing project to streamline the review process for civil cases, particularly in areas like personal injury and clinical negligence, The Law Society Gazette reported.
Noah Milton, AI project director for Fletchers, shared the progress at the Motor Accident Solicitors Society (MASS) conference in Manchester. Milton explained that the AI system has significantly improved since its inception nearly a year ago, already processing data from 70,000 cases. The system has proven particularly adept at flagging cases unlikely to succeed, reducing the workload for Fletchers’ legal team.
"When it says to reject a case, it is right 91% of the time," Milton told the conference audience. While AI is more effective at identifying claims that lack merit than spotting those that should be pursued, the tool is helping the firm make faster, more efficient decisions. "We are not trying to make a final judgment but to reduce the burden on our people," Milton said. "The tool is very good at identifying cases that are unlikely to be taken on, which frees up valuable lawyer time."
The AI development project, which began last year in collaboration with the University of Liverpool, focuses on structured information decision support systems (SIDDS). The goal is to reduce the time spent assessing clinical negligence cases and other claims, potentially saving thousands of staff hours. Fletchers has invested heavily in the four-year, multi-million-pound project, funded by its parent company, an affiliate of private equity investor Sun European Partners.
Milton also highlighted the firm’s ongoing work with generative learning AI tools that can review vast medical records. This technology has the potential to sift through disorganized and unstructured files, including handwritten doctor’s notes, to create a concise timeline of events relevant to a case. "Lots of clinical negligence cases hang on one page, and you can find that very quickly," Milton noted. Fletchers plans to begin testing this AI tool for personal injury cases in the coming months.
Fletchers Group CEO Peter Haden, who launched the project last year, emphasized the broader impact of AI in the legal sector. "We have already shown that AI can be a force for good," he said. "We now intend to broaden the scope of AI to give many thousands more injured people the opportunity to benefit from the justice system, even for lower-value cases where justice has been harder to access following government reforms and budget cuts."
