TL;DR:
In Fontainebleau Florida Hotel, LLC v. Botach, the Southern District of Florida on October 21, 2026 held that AI prompts and AI-generated outputs can be protected as work product, but protection is not automatic and must be evaluated document-by-document with a privilege log. The decision emphasizes that parties may seek protective orders and carefully tailor productions from AI providers, while courts scrutinize how AI was used in litigation. For trial teams, the ruling reinforces the need to document AI usage, segregate AI inputs from outputs, and prepare to defend or challenge AI-derived material at deposition and trial. Objection Academy offers practice modules to help litigators object to AI-related material, cross-examine experts using AI, and run realism-oriented simulations to prepare for AI-influenced evidence in court. This development signals a practical shift in AI discovery strategy, with continued attention from federal courts nationwide. (pacermonitor.com)
What happened and why it matters
On October 21, 2026, Magistrate Judge Marty Fulgueira Elfenbein issued a discovery order in Fontainebleau Florida Hotel, LLC v. Botach, a defamation matter pending in the U.S. District Court for the Southern District of Florida. The court held that work-product protection can apply to AI prompts and AI-generated outputs even when those materials were not created at the direction of counsel. The decision also mandated a privilege log for any AI materials that are withheld, and it addressed the process by which AI-provider productions should be handled so as not to bypass the litigant’s opportunity to assert work-product claims. This ruling is part of a broader gridlock among courts about how to treat AI-assisted materials in discovery, particularly whether inputs (prompts and guiding questions) must be disclosed alongside outputs that inform a party’s case. The order and its subsequent developments were reflected in docket activity, including a second order after a discovery hearing on October 28, 2026. (debevoise.com)
The Fontainebleau decision sits at the intersection of Rule 26(b)(3) protections and innovative uses of AI in litigation. The court’s analysis aligns with a growing view that AI-generated material may involve distinct layers of attorney thought and strategy, and that those layers deserve protection where appropriate, even as other AI-derived outputs may need to be produced. The opinion explicitly notes that a blanket rule is inappropriate, and it requires document-by-document evaluation and the furnishing of a privilege log for protected items. This approach marks an important strategic inflection point for trial teams weighing whether to rely on AI tools in case analysis, drafting, and experts’ work product. (debevoise.com)
Practical implications for trial teams
Documentation and privilege logs matter. The Fontainebleau order underscores the need to maintain a contemporaneous record of how AI was used in the litigation, including what prompts were issued and what outputs were generated. When possible, teams should produce materials that fall outside privileged work product and reserve sensitive AI-driven reasoning as protected. Counsel should prepare robust privilege logs that distinguish AI inputs (prompts, search strategies, and questions) from AI-generated outputs that may be discoverable. (debevoise.com)
Protect or limit training data exposure. The decision highlights the risk that AI tooling could train on sensitive inputs. Where feasible, employ enterprise-grade AI platforms that promise not to train on client data and to maintain input confidentiality, and include explicit guardrails in vendor agreements. This practice can strengthen privilege claims and reduce inadvertent waivers during discovery. (debevoise.com)
Tailor protective orders with AI-specific provisions. Courts increasingly expect thoughtful discovery orders that address AI usage. Protective orders may need to specify: (a) whether AI prompts or AI-generated notes may be produced or withheld; (b) the conditions under which AI tools may be used on sensitive materials; (c) safeguards to prevent training on disclosed materials. The Fontainebleau decision adds to a growing body of practice that can guide Rule 26(f) discussions and protective-order drafting. (debevoise.com)
Prepare for cross-examination and impeachment of AI-influenced opinions. Even where AI prompts or deposition outlines are protected, AI-derived outputs that informed an expert’s opinion may be the subject of discovery, and experts can be cross-examined about the reliability and provenance of AI inputs. Litigants should equip trial teams with robust cross-examination plans to challenge AI-assisted conclusions, including questions about data sources, training data, and the chain of provenance for AI-generated material. (debevoise.com)
Align trial-readiness training with evolving norms. As courts continue to adjudicate AI discovery issues, trial teams benefit from structured training that addresses both the admissibility and the reliability of AI-derived content. Objection Academy’s materials offer drills and simulations designed to sharpen objections to AI-derived evidence, enhance direct and cross-examination of experts relying on AI tools, and simulate courtroom scenarios involving AI-originated materials. This kind of practice helps lawyers stay fluent in the evolving discovery landscape. (objectionacademy.com)
How Objection Academy fits into this evolving landscape
Objection Academy provides targeted practice for trial attorneys to master objections, evidence handling, and trial-readiness skills in contexts that increasingly involve AI-informed litigation. The Villanueva decision in Nevada, for example, illustrates that AI-generated case summaries can be discoverable while AI prompts and deposition outlines may remain protected as drafts or work product; OA’s content reflects the practical need to prepare for both sides of AI-driven discovery disputes. OA’s resources help litigators:
- Craft precise objections to AI-derived content, addressing admissibility, foundation, and reliability concerns.
- Conduct rigorous cross-examinations of experts who may rely on AI tools, focusing on AI provenance and the extent of reliance.
- Run realistic trial simulations that incorporate AI-originated evidentiary materials, rehearsing strategies to mitigate hallucinations or miscitations.
- Build repeatable, objection-driven workflows that align with evolving discovery norms around AI.
These capabilities are designed to translate the courtroom realities created by Fontainebleau and similar decisions into concrete, repeatable practice for trial teams. (objectionacademy.com)
Next steps for practitioners
Assess and document AI usage in training and drafting. Create a policy for how experts and attorneys may use AI, and maintain logs showing what inputs were used and for what purpose. This supports precise responses to Rule 26 discovery requests and strengthens privilege claims where warranted. (debevoise.com)
Prepare for targeted AI discovery requests. Expect requests directed at AI case summaries, prompts, and AI-generated deposition outlines. Develop a plan for privilege logging, redaction, and protective-order arguments to preserve work-product protections where applicable. (debevoise.com)
Integrate AI-discovery readiness into Rule 26 conferences. Use the conference to set expectations for disclosure, establish a process for how AI materials will be produced or withheld, and memorialize agreed guardrails. OA-style practice can help teams rehearse these conversations and prepare effective responses. (objectionacademy.com)
Leverage OA for ongoing trial readiness. As AI-related discovery continues to evolve, regular training on objections, cross-examination of AI-reliant experts, and realistic simulations will help ensure courtroom fluency when AI material enters evidence. OA’s current materials reflect the emphasis on practical, repeatable drills that align with contemporary rulings. (objectionacademy.com)
Sources
Fontainebleau Florida Hotel, LLC v. Botach, 1:25-cv-20251, Southern District of Florida; Oct. 21 and Oct. 28, 2026 orders (Pacermonitor docket and public opinions). (pacermonitor.com)
AI Chats May Be Work Product—Even When Lawyers Aren’t Involved, Debevoise & Plimpton LLP, Oct. 28, 2026 (summary of the Fontainebleau order and its implications). (debevoise.com)
Nevada District Court Splits on AI Prompts in Expert Discovery: Villanueva v. LVMPD, Objection Academy, Sept. 29, 2026 (contextual relevance to AI prompts and outputs in discovery). (objectionacademy.com)
Federal Court Splits On Disclosure Of Expert Witness AI Prompts, Mayer Brown, Oct. 2026 (context on circuit splits and discovery practices). (mayerbrown.com)
Objection Academy, Nevada article page (OA analysis of Villanueva and practical steps for trials). (objectionacademy.com)
This timely development reinforces the practical reality that AI materials in litigation require careful handling, with concrete steps for privilege management, discovery planning, and trial preparation. Objection Academy remains a valuable resource for trial teams navigating these challenges, offering actionable training that aligns with the latest court practices and emerging AI governance in the courtroom.