The canvas is not dead
summary
Argues the canvas still matters when problems carry deep history and risk and shows what that means in practice: uses Yarn narrative framework plus HTML/JavaScript to assemble a web game that ships with 20 characters with distinct voices and hand‑drawn avatars; finds Figma and ChatGPT helpful for synthesis but Figma and ChatGPT fail at early visioning for a 700M‑user product where prompts create more rework than sketching; lays out three decision questions: context depth, expansiveness, and cost of being wrong.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 3:47AI in game development
- 8:46AI vs design process
- 13:47AI integration lessons
- 17:22Where AI excels and where it fails
- 20:54Accountability with AI
- 21:38Closing reflections
related talks
56mBeyond Chat: Bringing Models to the Canvas
15mA Canvas for Participation-Centered Game Design
06mInterview
25mUnexpected places: Figma as a space for anything and everything
25mCrafting your AI play(book)
28mEven by design
41mWeird Web & Curious Creation
18mThink You Can Build a Game with AI? Think Again!