Cover Image for The Unwritten Rules of AI Governance for Legal & Compliance Teams
Cover Image for The Unwritten Rules of AI Governance for Legal & Compliance Teams

The Unwritten Rules of AI Governance for Legal & Compliance Teams

Hosted by Pritesh Baviskar & Yuga Thombare
Google Meet
Registration
Approval Required
Your registration is subject to host approval.
Welcome! To join the event, please register below.
About Event

​When AI starts acting, who is responsible for it?

​THE SUPREME COURT HAD TO DEAL WITH A LEGAL CASE THAT AI HAD MADE UP.

​A team of lawyers had relied on AI-generated case law that did not exist, including a fictional judgment called “Mercy vs Mankind.” The Supreme Court has since flagged the growing problem of lawyers relying on AI-generated, non-existent judgments.

​That was AI assisting a lawyer.

​Now consider what happens when AI moves beyond assistance.

​An AI agent triggers a workflow.
AI communicates with a customer.
AI takes an action without someone approving every step.

​Who is responsible then?

​That is where AI governance gets much harder for Legal and Compliance.

​The questions we will tackle

​Who owns an AI decision?

​When Business, IT, Legal, Compliance and an external vendor are all involved, who is accountable when something goes wrong?

​What actually counts as human oversight?

​If someone reviews an AI recommendation and clicks “Approve,” is that meaningful oversight? What should a human be able to understand, challenge or override?

​Where does AI governance begin?

​Employee use of AI. AI embedded in software. Third-party models. Autonomous agents. Where does the organization's responsibility start and stop?

​When does an AI policy become a control?

​Who approves an AI use case? Who monitors it? What happens when the system, data or use case changes?

​Can you prove what happened?

​If an AI-related incident reaches Legal, Audit, a regulator or the board six months later, can you reconstruct what the AI did, what data it used, who approved it and what happened next?

​Real cases & questions.

​We'll use real cases and emerging AI developments to make these questions concrete, including AI-generated legal research, AI-generated customer communications, claims about AI capabilities, and the emerging question of liability when AI agents act autonomously.

​The goal is to think through what Legal and Compliance teams should be asking now, while AI is already making its way into everyday business processes.

​General Counsels | Legal Heads | Compliance Leaders | Risk Leaders | Senior In-house Legal & Compliance Teams

​30-minute discussion + 15-minute Q&A

​Come with the AI governance question your organization has not answered yet.

​Reserve your seat.