AI-Assisted Pro Se Filings Trigger Court Sanctions Amid Legal System Strain

Oscar Brownfield’s attempt to use artificial intelligence to bolster his pro se case against the Cherokee County School District backfired spectacularly. Representing himself in federal court in Oklahoma, Brownfield sought sanctions against his employer’s lawyers for allegedly filing false claims. But his AI-assisted motion cited completely fabricated cases, prompting the defense to pursue $7,000 in sanctions for their wasted time. A judge ultimately fined Brownfield $500 and warned of harsher penalties for future AI-related missteps.

Brownfield’s case is far from isolated. Courts across the country have recently flagged and sanctioned pro se litigants—workers representing themselves in federal labor and employment suits—who submit AI-generated briefs riddled with “hallucinations”: fictitious precedents and distorted legal citations. This surge in AI-assisted filings coincides with a rise in pro se lawsuits, as accessible chatbots like ChatGPT tempt individuals to navigate complex legal battles without lawyers.

Judges acknowledge AI’s potential to help non-lawyers better understand and respond to legal motions. Chief Magistrate Judge Vera Scanlon of New York noted that some AI-generated responses “seemed to make a lot of sense” until closer review exposed nonexistent cases. The extra scrutiny required is stretching court resources and frustrating defense attorneys, who must sift through voluminous, inaccurate filings.

Legal experts see a paradox: AI could democratize justice in a system ranked 112th globally for civil justice accessibility by the World Justice Project. Yet the technology’s current shortcomings risk undermining that promise by introducing errors that courts must police. Some law schools are training low-income individuals on AI tools to improve pro se navigation, but the technology still demands robust governance.

Defense lawyers complain that AI encourages pro se litigants to file excessive, complex submissions, prolonging litigation and increasing costs. Courts vary in their responses—some ban AI use outright, others require filers to certify accuracy. Sanctions remain a blunt instrument, as judges balance deterring misuse against preserving access to courts for those without lawyers.

Beyond courtroom filings, a new legal battleground is emerging over discovery of AI interactions. Recent rulings differ on whether communications with chatbots like ChatGPT or Anthropic’s Claude are protected by attorney-client privilege or work-product doctrine, especially when pro se litigants are involved. The absence of clear national standards promises more litigation over AI’s role in legal strategy.

On the ground, attorneys report shifting client relationships as AI reshapes expectations. Some clients want limited representation, relying heavily on AI drafts they ask lawyers to review. Lawyers are adapting contracts to address AI use and educate clients on its limits, striving to maintain trust in human legal expertise.

Oscar Brownfield’s $500 fine is a cautionary tale of AI’s double-edged sword in the courts. While it can empower workers to fight back, hallucinated legal claims threaten to clog the justice system and invite sanctions. The future of AI in legal self-representation hinges on better tools, clearer rules, and a reckoning with how technology intersects with access to justice.