Ross Lovegrove’s AI-designed chair shows where human judgement still matters

Designer Ross Lovegrove used Google DeepMind’s Gemini in an experiment that generated hundreds of variations on a chair, but the chosen design proved heavy and not entirely ergonomic when made physically. The project is a useful glimpse of AI as a creative amplifier, not a button marked “finished product”.

Google DeepMind Watch analysis

What happened

In an interview published by Parametric Architecture, Lovegrove describes introducing his drawings and design language into an AI-driven process using Gemini. The resulting chair variations did not all reflect his style, and turning the selected design into a physical object exposed practical problems.

Lovegrove presents the experiment as a negotiation between computational generation, authorship, materials, ergonomics and manufacturing. He describes the potential as especially interesting when AI meets a designer’s accumulated experience and body of work. Read the interview.

Why it matters

Generating hundreds of forms is the easy-to-count part. Knowing which are worth making, and whether they work as chairs rather than persuasive images, takes judgement and engineering. Lovegrove’s example puts that gap in view without pretending that every design experiment needs to end in a perfect object.

It also points to a broader question for creative tools: might AI make a distinctive designer’s instincts more productive, rather than make expertise less necessary? That is a more interesting prospect than another gallery of machine-generated furniture with nowhere comfortable to sit.

Our read

This is a telling case study, not evidence that AI has solved product design. The useful lesson is to judge creative systems by what survives contact with materials, bodies and production, not by the number of variations they can produce. Designers should treat the outputs as material for decisions, not decisions in themselves.

What to watch

• Whether later versions of the chair address its weight and ergonomic shortcomings.
• How designers turn generative variations into manufacturable objects.
• Whether AI tools become easier to direct through sketches, language or other creative interfaces.

Discussion spark: When AI generates design options, should the main measure of success be how many ideas it produces, or how well designers can turn the best ones into usable objects?

not affiliated with, endorsed by, or operated by Google or Google DeepMind