The jagged technological frontier describes the uneven edge of what AI does well: tasks of similar difficulty can fall on either side of it, and the line is hard to spot in advance.

Where it comes from

The term comes from a 2023 field experiment by researchers from Harvard, Wharton, MIT and other institutions with 758 consultants at Boston Consulting Group.¹ On tasks inside the frontier, consultants using GPT-4 completed 12 percent more tasks, 25 percent faster, at quality more than 40 percent higher. On a task just outside it, those using AI reached the right answer less often than colleagues working without it. The authors also described two ways of working with AI: "centaurs," who split tasks between themselves and the machine, and "cyborgs," who weave AI into every step.

Where it goes wrong

The frontier is often treated as a fixed map. It moves with each new model and differs by task and by organization, so every team has to keep rediscovering where it runs.

Talent Takeaway

Keep a shared team list of where AI has proven reliable and where it has let you down, and review it whenever your tools change. The list itself becomes a learning tool for new colleagues.

More on the study: AI Made Consultants 40% Better, With One Catch

¹ Dell'Acqua, F., McFowland III, E., Mollick, E. R., Lifshitz-Assaf, H., Kellogg, K., Rajendran, S., Krayer, L., Candelon, F., & Lakhani, K. R. (2023). Navigating the jagged technological frontier: Field experimental evidence of the effects of artificial intelligence on knowledge worker productivity and quality (Harvard Business School Working Paper No. 24-013). https://doi.org/10.2139/ssrn.4573321

Further reading: Mollick, E. (2024). Co-intelligence: Living and working with AI. Portfolio.