AI floods courts with self represented lawsuits

AI is driving a flood of self represented lawsuits.

A sweeping look at federal civil cases shows a sharp uptick in filings by people who appear without lawyers, and the trend appears tightly linked to how AI tools are used in paperwork and drafting. The team behind the study analyzed 4.5 million federal civil cases spanning 2005 to 2026 and finds the share of filings by self represented litigants rose from 11% in 2022 to 16.8% in 2025. Even more striking, filings by unrepresented litigants have more than doubled since pre-2023 levels. The authors argue that AI-enabled drafting and document processing are a meaningful driver of that rise, even as access to justice improves in some cases by lowering the barrier to filing.

In real courtroom terms, the trend is being felt in judges’ chambers as well as on the docket. Judge Maritza Braswell, a federal magistrate judge in Colorado, says she sees AI in every corner of the workflow. She uses AI to vet court documents and says she can spot when AI generated prose shows up in pleadings, including hallucinated quotes. “I correlate the jump to AI in part because I see AI use,” she says. Yet she also notes a paradox: the same AI that helps people file more easily does not translate into a higher win rate for self represented litigants. The surge in filings comes with a parallel question about accountability for AI when it gives bad legal advice or fabricates quotes, a concern lawmakers are only beginning to grapple with across states.

The study’s findings put a sharper spotlight on the practical consequences for courts that already stretch thin. The rise of AI assisted drafting is cited as a factor behind better drafted pleadings, even as judges question how to assign responsibility for the quality and accuracy of those pleadings when the author is a machine. The broader policy debate is now moving beyond whether AI is useful to who bears the cost when AI produced guidance goes wrong. Lawmakers are weighing questions about liability for AI generated advice and the extent of a litigant’s responsibility to verify what an AI assistant returns.

From a practitioner standpoint, the moment exposes both promise and peril. On the one hand, AI powered tools can help those who cannot afford counsel to initiate claims that might otherwise stagnate. On the other hand, the same tools can generate content that looks sophisticated but contains faulty logic or misquotes, creating downstream risks for judges and for the unrepresented parties who rely on them. The court system will likely need guardrails, such as clear standards for when human review is required, guidance on permissible AI generated content, and funding for training and oversight, to prevent these tools from widening inequities or introducing new biases into proceedings.

Looking ahead, observers say two practical trends will shape the next year. First, more courts will experiment with AI assisted triage of filings, flagging potential issues for human review rather than replacing it. Second, the policy conversation will intensify around who pays when AI advice misleads, and how to design governance that preserves access while guarding against harmful mistakes. In the near term, the trend suggests a continued rise in self represented filings, driven in part by AI assisted drafting, with judges balancing the potential for greater access against the need for accuracy and accountability in the courtroom.