Apprenticeship by chance, development by design

The American Lawyer wrote in a post dated 2026-09-03 that firms making meaningful investments in well-designed AI workflows can replace apprenticeship by chance with development by design. Wayne Stacy and Caren Ulrich Stacy wrote the formulation. It is the clearest development in the material: AI is presented as a way to change how lawyers are developed, rather than merely as a way to reduce the amount of beginner work.

That distinction matters because the advantage belongs to firms that grow lawyers faster. A workflow that deliberately exposes lawyers to useful work, feedback, and increasing responsibility could support that advantage if it actually produces capable lawyers. The American Lawyer connects AI investment to that outcome, while Daniel W. Linna Jr. says AI can help students learn more, faster through tutors and simulations. Together, the claims point toward development as the valuable output of AI, rather than headcount reduction as the output that matters.

The stronger reading remains unproved. The American Lawyer names no firm, workflow, matter, lawyer, training period, or measured change in capability. Its statement says what well-designed workflows can do, but it does not show that a firm has built one that develops lawyers faster. Daniel W. Linna Jr. describes possible tools for students, not a legal workplace result. The distinction is decisive: a design for learning is not the same thing as demonstrated lawyer development.

Other formulations support the direction without closing the gap. Sarah Guo says AI may create a country of geniuses in a datacenter by combining aggregate intelligence, continuous operation, and greater communication bandwidth. She also relays an engineer’s view that AI has been replacing a job while leaving people busier and able to pursue more ambitious work. Those claims suggest that AI can expand the amount or reach of human work. They do not establish that legal workers acquire judgment faster, that responsibility moves safely to them, or that a firm produces stronger lawyers as a result.

The next confirming fact would be a firm naming an AI workflow and publishing a before-and-after measure of lawyer capability, such as the work lawyers can handle independently or the time required to reach that point. A launch alone would show deployment; a measured change in lawyer development would support the stronger claim. Until then, the event supports development by design as a plausible direction, while the advantage for firms remains contingent on proving that lawyers actually grow faster.