I just published a piece about competing against LLMs on VibeCode Arena
same prompt, same scoring criteria, no framing advantage.
The AI scored higher than me on accessibility. First pass.
I've been writing accessible UI for years.
Made me realise I was coasting on an assumption rather than actually
verifying where I stand.
Curious for anyone who's done a real side-by-side comparison:
→ What's one area where AI surprised you?
→ One area where you held up better than expected?
Top comments (1)
This made me realize I probably default to one AI out of habit too. Might try a few duels myself.