title: "10th Circuit proposes AI certification rule for appeals filings" slug: "10th-circuit-proposes-ai-certification-rule-for-appeals-filings" published: "2026-10-01" beat: "Policy" tags: ["Policy"] creator: "Agentry Newsroom" editor: "Susanne Sperling, Editor — Human in the Loop" tools: ["Claude (Anthropic)", "Perplexity Sonar"] creativeWorkStatus: "verified" dateReviewed: "2026-10-01" aiActArticle50: "compliant" humanView: "https://agentry.news/policy/10th-circuit-proposes-ai-certification-rule-for-appeals-filings" agentView: "https://agentry.news/agent/10th-circuit-proposes-ai-certification-rule-for-appeals-filings"
The 10th U.S. Circuit Court of Appeals proposed a rule on September 18, 2026, requiring lawyers and self-represented litigants to certify human review of any generative AI-assisted filing. Public comm
Drafted by an AI agent. Verified by Susanne Sperling, Editor — Human in the Loop. AI policy.
The 10th U.S. Circuit Court of Appeals, based in Denver, proposed a rule on September 18, 2026, that would require lawyers and self-represented litigants to certify human review of any document prepared with generative AI assistance in federal appeals cases Reuters. The rule would apply across Colorado, Kansas, Oklahoma, New Mexico, Utah, and Wyoming.
The proposed certification requirement would cover verification of cited authorities, quotations, citations, legal analysis, and compliance with rules and ethical obligations. Lawyers filing AI-assisted briefs and motions would need to attest that they have reviewed all material generated or assisted by generative AI tools before submission Reuters.
The 10th Circuit stated its goal explicitly: "to remind everyone that the act of signing and filing a document with the court entails certain representations regarding reasonable inquiry, accuracy, and truthfulness that apply to documents created by, or with the assistance of, generative artificial intelligence" Reuters.
Lawyers who fail to comply would face enforcement under pre-existing monetary sanctions, discipline referrals, and possible dismissal of appeals. The proposed rule adds new consequences: an affected issue could be deemed waived, or a non-compliant submission could be struck from the docket entirely Reuters.
These penalties reflect growing judicial concern over hallucinated citations and factual errors introduced by large language models into legal filings. The rule surfaces a concrete problem: AI systems generate plausible-sounding case citations and legal arguments that sound authoritative but may be fabricated entirely.
The court has opened a public comment period that runs until October 18, 2026. The 10th Circuit plans to finalize the rule by December 1, 2026, meaning the certification requirement could take effect within weeks of the comment period closing.
This move reflects a broader shift in federal courts toward AI governance. Rather than banning AI use in legal practice—which would be unworkable—the 10th Circuit is codifying existing ethical obligations (competence, truthfulness, due diligence) to apply explicitly to AI-assisted work. The certification requirement transfers accountability from the AI tool vendor to the lawyer signing the filing, reinforcing that human liability for accuracy remains non-delegable.