A UX Design article argues designers should stop talking AI agents through design decisions in chat and bottle their judgment into a structured DESIGN.md file instead.
JetBrains releases the Kotlin Benchmark, evaluating AI coding agents on end-to-end Kotlin tasks from issue triage to validated solutions.
AI coding tools like Cursor and v0 generate components that ignore design tokens, shifting design system debt from designers to engineers. Three resources worth reading.
New research from MIT shows LLMs assign roles based on writing style, not role tags. Prompt injection works because untrusted text can simply sound like a system prompt.
Daniel Stenberg says AI-assisted security reports hit curl at 4-5x the 2024 rate. The quality is higher, the volume is unsustainable, and his wife is worried.
Developer David Wilson catalogued 16+ unfinished AI-assisted projects and concluded the fix might be cancelling his subscription entirely.