AI boosts junior patent attorneys but limits long-term skill retention

A three-month randomized controlled trial involving 133 patent attorneys across eleven U.S. intellectual property law firms examined the impact of AI assistance on professional performance. The Google-funded study provided two-thirds of participants with access to InFlow, an unreleased AI assistant from Google Labs, while the remaining one-third served as a control group without AI support.

During drafting tasks, the group using AI outperformed the control group at both the 10-day and 90-day marks across all five scoring dimensions. Performance gains were significantly larger for junior attorneys with fewer than seven years of experience compared to their senior counterparts. However, when the AI was removed from the workflow, only senior attorneys retained the performance benefits gained during the experiment.

In a separate redlining task where all participants reviewed drafts without AI, the advantage was entirely driven by senior attorneys. Junior attorneys saw no benefit in this task despite having posted the biggest gains during AI-assisted drafting. The study noted that junior attorneys frequently spotted serious flaws without fixing them, leaving notes that read like instructions to an AI assistant, whereas seniors more often restructured flawed drafts and grounded changes in legal principles.

The authors suggest that baseline expertise may be a prerequisite for turning AI-assisted practice into durable ability. This finding aligns with other research, including an Anthropic study showing younger developers scored worse on follow-up tests after learning with AI, and a BCG study indicating weaker consultants improved more with GPT-4 than strong performers. Anthropic also reported a drop in junior engineering positions since early 2025.