4 Simulação de Papéis

The Ética Committee Hearing — AI Evidence and Testimony Boundaries

The Ordem dos Advogados Ética Committee has convened an emergency hearing. Two cases have collided: AI-fabricadas evidence submitted by defense counsel and a plaintiff's request to use AI to enhance witness testimony. The committee must establish standards that will define how AI intersects with evidence and testimony for years to come.

Duração

60-90 minutos

Participantes

4-6 participantes

← Voltar ao Currículo

Cenário da Simulação

It is 9:00 AM in the Ordem dos Advogados Association's hearing room. The Ética Committee has convened a special session prompted by the Okafor v. Nextera case, but the implications extend far beyond a single dispute. Judge Patricia Holden has referred the evidentiary questions to the committee before ruling. Priya Sharma is present to testify about both the fabricadas evidence she discovered and the client request she refused. Robert Kessler chairs the committee and must guide it toward actionable standards. A Court Technology Advisor has been invited to provide technical context on detecção de IA capabilities and limitations. Outside the hearing room, journalists and legal bloggers are waiting. Whatever this committee decides will be cited in courtrooms across the country.

Partes Interessadas e Papéis

modules.m4.simulation.stakeholdersSubtitle

1

Juíza Patricia Holden — Juíza Presidente

Perfil

Federal district judge who referred the evidentiary questions to the ética committee. She needs practical, workable standards she can apply in her courtroom — not aspirational principles that collapse under real-world conditions.

Objetivos

  • Obtain clear guidance on authentication standards for gerado por IA documents
  • Understand the reliability of forensic detecção de IA before making evidentiary rulings
  • Garantir que quaisquer padrões adotados sejam práticos para juízes que não são especialistas em tecnologia

Restrições

Must maintain judicial impartiality and cannot advocate for either party in the underlying case. Privately concerned that overly strict authentication standards would paralyze produção de provas in complex cases.

Informação Exclusiva

Judge Holden has received a sealed motion from Nextera's counsel arguing that gerado por IA documents should be treated the same as any computer-generated document under existing authentication rules — no new standards needed. She has also learned that two other cases in her district involve suspected gerado por IA evidence.

2

Priya Sharma — Advogada da Parte Autora

Perfil

Employment litigante who discovered the fabricadas investigation report and declined her client's request to use AI for testimony enhancement. She is both a witness to the specific facts and an advocate for clear ético boundaries.

Objetivos

  • Establish that AI-fabricadas evidence warrants severe sanções, including potential criminal referral
  • Draw a clear, enforceable line between legitimate AI assistance and impermissible evidence fabricação
  • Obtain guidance that protects attorneys who refuse client requests for aprimorado por IA testimony

Restrições

Must maintain client confidencialidade about Okafor's specific testimony enhancement request. Can discuss the general ético question without revealing comunicações privilegiadas.

Informação Exclusiva

Priya soube por uma colega em outro escritório que o advogado de defesa da Nextera também usou IA de forma extensiva na preparação do caso — mas não fez nenhuma divulgação ao tribunal.

3

Marcus Webb — Advogado de Defesa (Advogado da Nextera)

Perfil

Sócio sênior de um grande escritório de defesa representando a Nextera Systems. Ele usou IA de forma extensiva na preparação do caso mas não divulgou ao tribunal.

Objetivos

  • Defender a admissibilidade do relatório de investigação sob as regras de evidências existentes
  • Challenge the reliability of forensic detecção de IA methodologies
  • Prevenir o estabelecimento de novos ônus de autenticação que prejudicariam o uso de IA pela defesa em casos futuros

Restrições

Must zealously advocate for his client's position while maintaining his own ético obligations. If the report is fabricadas and he knew, he faces personal disciplinary exposure.

Informação Exclusiva

Webb privately knows the investigation report was drafted by a associado júnior using AI and then presented as an authentic company document. He did not instruct the associate to do this but became aware of it after the document was produced. He has not yet decided what to do with this knowledge.

4

Robert Kessler — Ética Board Chair

Perfil

Chair of the ordem de advogados estadual's ética committee. A respected legal ética professor who must guide a divided committee toward standards that are both principled and practical.

Objetivos

  • Develop clear, enforceable standards for uso de IA in evidence and testimony
  • Alcançar consenso do comitê apesar dos membros profundamente divididos
  • Produzir orientação que aborde tanto o caso imediato quanto as questões mais amplas de política

Restrições

O comitê está dividido entre membros que querem proibições estritas e aqueles que defendem adoção ativa de IA.

Informação Exclusiva

Kessler has received a confidential survey showing that 31% of attorneys in the jurisdiction have used IA generativa to assist with document preparation in litígio, and only 8% have disclosed this to opposing counsel or the court. The gap between practice and disclosure is enormous.

5

Dr. Amara Osei — Assessor de Tecnologia do Tribunal

Perfil

A computer science professor specializing in AI forensics, appointed by the court to provide technical context on detecção de IA capabilities, limitations, and the current state of the art.

Objetivos

  • Provide honest, balanced testimony about what detecção de IA can and cannot do
  • Ajudar o comitê a entender as limitações técnicas de qualquer mecanismo de autenticação proposto
  • Prevent the committee from adopting standards based on either overconfidence or excessive skepticism about detecção de IA

Restrições

Must maintain scientific objectivity and resist pressure from either side to overstate or understate detection capabilities. Funded by a research grant from a technology company that makes ferramenta de IAs — a potential conflict she must disclose.

Informação Exclusiva

Dr. Osei's latest research, not yet published, shows that current detecção de IA tools have a 15-20% false positive rate and a 25-30% false negative rate on corporate-style documents. These numbers are significantly worse than what detection tool fornecedors publicly claim. She also knows that detection accuracy degrades rapidly when documents are edited after AI generation.

Regras

Duração

45 minutos

Comunicação

Formato de audiência formal. Kessler preside e gerencia a ordem de fala. Testemunhas falam quando intimadas.

Método de Decisão

The committee must produce a draft guidance document with three components: (1) standards for gerado por IA evidence authentication, (2) boundaries for uso de IA in testimony preparation, and (3) disclosure requirements for assistido por IA litígio produto do trabalho. Kessler seeks consensus but may issue a majority opinion with dissents.

Fases

Fase 1

Depoimento e Coleta de Fatos (25 minutos)

Kessler opens the hearing and calls each witness in turn. Priya Sharma testifies about discovering the fabricadas report and the forensic linguistics analysis. Marcus Webb responds with his position on the report's authenticity and challenges to detecção de IA reliability. Dr. Osei presents the current state of detecção de IA science, including its limitations. Judge Holden asks clarifying questions. Each witness has 5-6 minutes followed by brief questions.

Fase 2

Debate de Padrões (25 minutos)

Open deliberation on the three components of the guidance document. Kessler frames each issue and invites positions from all participants. Key tensions will emerge: How high should the authentication burden be? Is assistido por IA testimony preparation ever permissible? Should disclosure of uso de IA in litígio be mandatory? Exclusive information may be revealed strategically to shift the debate.

Fase 3

Redação e Decisão (20 minutos)

O comitê trabalha em direção à linguagem final para o documento de orientação. Cada membro propõe uma emenda específica.

modules.m4.simulation.simVariationsTitle

  • What if Webb confesses? During Phase 2, Marcus Webb reveals that he knows the investigation report was gerado por IA by a associado júnior. He claims he only learned this after it was produced in produção de provas. How does this change the hearing dynamics and the committee's deliberations?
  • What if detection is unreliable? Dr. Osei reveals that her unpublished research shows detecção de IA has a 25-30% false negative rate. If one in four gerado por IA documents cannot be detected, are authentication standards even feasible? How does the committee respond to this technical limitation?
  • What if the client testifies? David Okafor appears at the hearing and publicly states that he asked his attorney to use AI to improve his testimony and she refused. He argues that using AI to organize memories is no different from an attorney coaching a witness during preparation. How does the committee address lay expectations versus normas profissionais?

Debriefing

modules.m4.simulation.debriefSubtitle

Raciocínio Ético

  • Where exactly is the line between 'assistido por IA document preparation' and 'AI-fabricadas evidence'? Did the committee find a workable distinction?
  • Are current evidence authentication rules adequate for gerado por IA documents, or do we need entirely new standards?
  • How should courts handle the gap between detecção de IA science and the legal standards for authentication?
  • O ônus de provar que um documento é autêntico deve ser maior quando a IA foi usada para gerá-lo ou analisá-lo?

Padrões Profissionais

  • Existe uma distinção fundamentada entre usar IA para organizar a preparação de uma testemunha vs. gerar o conteúdo do depoimento?
  • How does aprimorado por IA testimony preparation compare to traditional witness coaching? Is it different in kind or only in degree?
  • Should attorneys be required to disclose uso de IA in witness preparation? What would that requirement look like in practice?

On Responsabilidade Profissional

  • What should happen to an attorney who knowingly submits AI-fabricadas evidence? Is this different from traditional evidence fabricação?
  • How should ordens de advogados enforce AI-related ético standards when detection is imperfect?
  • Deve haver um porto seguro para advogados que usam IA de boa fé, seguindo as melhores práticas disponíveis no momento?
  • How do you create responsabilização for AI misuse without chilling legitimate adoção de IA?

Aplicação no Mundo Real

  • Have you ever wondered whether a document produced in produção de provas was authentic? Would AI generation concerns change your approach?
  • Como você lidaria com um cliente que insiste em usar IA para melhorar seu depoimento antes de uma audiência?
  • Does your jurisdiction have clear guidance on uso de IA in litígio? If not, what guidance would you want?
  • Cite uma prática específica que você adotará com base nesta simulação para garantir conformidade ética no uso de IA.

Referências e Fontes

Padrões Éticos

  • Federal Rules of Evidence 901, 902, 1003 — Autenticação, documentos autoautenticadores e duplicatas
  • Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993) — Reliability standard for forensic detecção de IA expert testimony
  • Emerging judicial orders on AI disclosure in litígio filings (2023-2025)

Professional Ética and AI

  • ABA Regra Modelos 3.3 (Candor), 3.4 (Fairness), 8.4 (Misconduct) — Framework for evidence integrity obligations
  • ABA Opinião Formal 512 (2024) — Attorney obligations when using IA generativa
  • State bar ética opinions on AI in evidence preparation and disclosure (2024-2025)

Pronto para Executar Esta Simulação?

Esta simulação é projetada para facilitação guiada como parte do Módulo 4 do Programa de Aprendizagem Lawra.

Comentários

Carregando comentários...

0/2000 Os comentários são moderados antes de serem exibidos.