Keep the human in the loop

An interesting article about developer reliance on AI worries IT leaders highlights the changing nature of "work" in an AI-led world. I'd summarize this article as: IT leaders are still looking for AI to streamline work, but there needs to be a human in the loop of that process.

The article says "the engineers who will thrive in this new era are adopting AI to enhance their thinking and output, not replace it." And I believe this is an important distinction: organizations that can deploy AI to multiply their capacity will lead in the new AI era.

But AI isn't here to replace human workers; AI is a tool, not an end goal. And this is also commented in the article, to some extent. The article mentions that nearly one-third of surveyed IT leaders consider overreliance on AI as a top concern. Almost all (95%) remain cautious about risks when using AI-generated code.

This is the human in the loop that I referred to earlier. The term is not new; according to Wikipedia, "human in the loop" has been applied to machine learning, simulation, and other systems. When referring to AI, human in the loop means developers must review AI-generated code, technical writers must edit and revise AI-generated text, and staff must maintain oversight on AI-controlled systems.