Prompt Engineering: Are Speaking Machine Again?

Rethinking UX in the Age of AI

For decades, UX design has helped translate the language of machines into something humans could understand. From command lines to graphical interfaces, each major shift in computing has reduced the burden on users and brought technology closer to how people naturally think and work.

But with the rise of AI, something interesting is happening, despite all our progress in human-centered design, we may have quietly returned to something that looks surprisingly familiar: the command-line era.

Artificial intelligence has rapidly moved from a niche capability to something embedded in our everyday workflows. Designers, developers, researchers, and analysts now use AI tools daily to generate ideas, summarise information, and accelerate tasks. Much of the conversation around AI focuses on efficiency. AI can streamline work and automate repetitive tasks by simply writing prompts. But that may only be the first layer of its potential.

The real opportunity lies in collaboration.

AI is beginning to behave less like a traditional tool and more like a thinking partner, assisting with problem-solving, ideation, and discovery. We’re already seeing this in fields like medicine and biology, where AI systems have helped researchers solve complex problems such as protein folding, opening new possibilities for drug discovery and treatment.

But this raises an important UX question:

"Are we interacting with AI in the best possible way for this type of collaboration?"

---

A Familiar Pattern in Technology

Looking at the history of computing, a clear pattern emerges:

Early computers required people to adapt to the machine. In the MS-DOS era, users had to write precise command-line instructions. It was powerful, but far from intuitive. Then came the graphical user interface (GUI): icons, windows, and buttons that allowed people to interact with computers in a more natural and visual way. This shift transformed computing from a specialist skill into something accessible to millions of people.

It wasn’t just a technical improvement. It was a UX revolution.

Today’s interaction with AI feels oddly similar to that earlier era. Much of our communication with AI relies on prompting. We write instructions, adjust phrasing, and experiment with different ways of asking for what we want. While prompting uses natural language, it still requires users to think about how the system interprets instructions.

Many users may find themselves wondering:

• Am I writing this correctly?

• Did I give enough context?

• Is there a better way to phrase this?

  
  

From a UX perspective, this means prompting still carries a significant cognitive load.

In many ways, prompt engineering resembles a softer version of the command line — powerful, but still requiring users to adapt their thinking to the system.

---

Prompting May Be a Transitional Interface

If history is any guide, prompting may simply represent an early stage of AI interaction. Just as command-line interfaces eventually evolved into graphical interfaces, AI interactions may evolve into more intuitive forms of collaboration, ones where users no longer need to carefully craft instructions. Instead, systems may better understand context, intent, and workflow, reducing the intellectual effort currently required.

We’re already seeing hints of this shift:

• In software development, GitHub Copilot enables AI-assisted coding, where developers effectively work alongside an AI that suggests solutions based on context.

• In design, tools like Figma are introducing AI features that help generate layouts, explore design ideas, and accelerate prototyping. Rather than replacing designers, the AI acts as a creative assistant.

• In science, systems like AlphaFold have helped researchers solve the decades-long challenge of predicting protein structures, enabling faster breakthroughs in biology and medicine.

  
  

These examples point to something bigger. AI’s real power may not lie in automation alone, but in co-creation between humans and intelligent systems.

---

The Next UX Opportunity

For UX designers and strategists, this moment presents a fascinating challenge:

• The question is no longer simply: How do we design AI tools?

• Instead, we must ask: How do we design meaningful human–AI collaboration?

  
  

History suggests that the biggest breakthroughs rarely come from the technology itself. They come from rethinking how humans interact with it.

If the command line defined early computing, prompting may define the early years of AI. But the real UX breakthrough will likely come when we move beyond prompting entirely, toward interfaces that allow humans and AI to work together more naturally.

That shift may represent the next great UX revolution. And as designers, we have an opportunity to help shape it.