The legal industry has spent the past two years pursuing new AI tools, but many lawyers haven’t changed how they work. The next phase of legal innovation should focus on optimizing AI-enabled workflows, says iManage adviser Ivy Grey. The next phase of legal innovation should focus on optimizing AI-enabled workflows for attorneys, says iManage adviser Ivy Grey.
Building Accountability into AI Workflows
A workflow is often seen as a technical diagram with steps and building blocks for automation. However, the key to an effective workflow lies in the proper order of these blocks and the handoffs between lawyers, associates, software, and AI agents. Identifying the building blocks is just the first step; the real challenge is ensuring they are sequenced correctly and that the transitions between them are seamless. This is where “accountability by design” becomes key for AI adoption and improving work processes.
Trusting a person or AI to act on your behalf requires building accountability into the natural course of work, such as confirming facts and citations during drafting or reviewing. Without this, the risk of errors or missteps increases, undermining the very efficiency AI aims to provide.
For instance, lawyers routinely tell a colleague or an AI tool to “just do X” without mentioning the 25 steps that have to happen first. This lack of clarity can lead to misunderstandings and inefficiencies, as the recipient may not fully understand the task’s scope or requirements.
Related Post: The Future of Case Lawyer: Navigating Tomorrow’s Legal Landscape
The Shift from Traditional to AI-Era Accountability
In the past, accountability was often an afterthought, added at the end of a process through audits and documentation. This approach is no longer viable in the AI era. The speed and complexity of AI-driven tasks demand real-time accountability, not post-hoc reviews.
Consider a litigator drafting a brief’s factual background, including a mortgage payment of $3,510 on a specific date. In the paper era, this required minimal scrutiny, as the physical check existed. With AI drafting, the amount, sender, and date need confirmation against the source, ideally with a link to the document. This additional layer of verification ensures accuracy and builds trust in the AI’s output, which is key for legal work where precision is non-negotiable.
Accountability and Efficiency in AI-Assisted Legal Work
The same discipline applies to legal requirements. Confirming that the analysis and application of lien law is correct remains a professional obligation, for example, a junior associate should check the facts and mark them as verified by inserting their initials. This allows senior lawyers to trust the associate’s work and continue from there. This structured approach not only ensures accuracy but also supports a collaborative environment where responsibilities are clearly defined and executed.
Best Practices for Workflow Optimization
A good handoff spells out the goal, the sources typically needed, and how to make a judgment call if a question arises. This distinction between delegation and assignment is key. Delegation provides enough context for successful task completion and judgment, while assignment leaves it to chance. Delegating a task means handing it off with enough context to do it well, and a clear enough picture of “done well” to judge the result when it returns.
