OpenAI's head of Codex says AI still can't get creative design right, pointing to the subjective nature of good taste as a hurdle that code doesn't face. Speaking on a recent episode of "Lenny's Podcast," Andrew Ambrosino explained why design remains harder for AI to master than software engineering, a reminder that creative roles still require a human touch.
"I think design's a little bit harder to grade than software," Ambrosino said. "Creating a loop where you can train the model on what's good design and what's bad design is just a little bit more tedious and onerous than, you know, does the code compile?" The comment reflects a longer-standing difficulty: AI can generate images and layouts, but evaluating whether a design works - whether it's truly appealing or effective - requires a level of subjective judgment that's hard to teach a model.
AI design tools produce 'average' results
Figma CEO Dylan Field made a similar point on an episode of "Hard Fork," arguing that AI models are trained on the "distribution of data" and tend to output designs that people recognize as average. That means they often land in a safe, unremarkable middle ground rather than pushing boundaries. For working designers, the takeaway is that AI serves best as a tool for exploration and iteration, not as a replacement that delivers finished, distinctive work.
Human taste remains the differentiator
Grammy-nominated musician Bas captured the same idea at a Harvard conference in 2024. "At the end of the day, there's going to be a certain level of human taste that's still required to prompt the AI to create something that we'll all enjoy," he said. The quote underscores that even the most advanced AI output still depends on a person to decide what's good - a process that isn't going away soon. Ambrosino echoed that sentiment with his closing line: "Let's give it up for the human brain for now."
Why this matters for creatives
The message from Ambrosino, Field, and Bas is consistent: AI isn't replacing creative judgment - it's changing how creatives work. The designers, artists, and musicians who thrive will be those who treat AI as a collaborator, not a threat, and invest time in learning how to prompt and refine its output. Resources like AI Design Courses & Certifications and AI for Creatives Courses & Certifications offer a starting point for building practical AI skills without losing the human perspective that makes design work.
Your membership also unlocks:
