Faster Horses, Not Trains (blog.robbowley.net)

A recent reflection on generative AI (GenAI) suggests that while advancements in models are occurring, they often feel less impactful to users who have integrated these tools into their daily workflows. The key argument is that GenAI models, despite becoming more sophisticated, still rely on a lossy interface when translating complex, real-world tasks into simplified digital formats. This inherent limitation means that GenAI is primarily effective in areas where data already exists in digital form, leading to incremental improvements rather than transformative changes in broader systems. The significance of this perspective lies in understanding GenAI’s current role within the landscape of technological advancement. It functions as an optimizer within existing constraints rather than a disruptor on the scale of historical innovations like the steam engine or electricity. While the potential for GenAI to drive major breakthroughs exists—especially if it can contribute to foundational areas like energy, food production, and material science—the path to such industrial-scale impact remains uncertain. Until these fundamental constraints are addressed, the progression of GenAI may continue to feel more like enhancing "faster horses" rather than pioneering “trains” that transform entire systems.