Glimpses of Humanity: My Game-Building AI Pair
summary
Builds a text‑driven escape‑room adventure called Awaken in Rails, pairing over 140,000 words (≈450 printed pages) with Atheniel‑née‑ChatGPT (ChatGPT‑4). Shows concrete choices: models for rooms and puzzles, session hash vs a game state model, an inventory system, a stimulus typewriter controller and Turbo morphing for partial page updates. Demonstrates surprising outcomes: the LLM drives play by proposing commands while the developer inputs responses, helps unblock design and presentation work, and sparks an unexpectedly human emotional moment.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 1:03Talk excitement
- 2:30AI and humanity fears
- 3:11Understanding large language models and interaction
- 7:53AI as teammate
- 10:03Naming Aenial
- 11:51Planning the game
- 16:09Game named Perchance
- 18:56Awaken overview
- 19:17Technical questions
- 21:26Development and pairing
- 22:12Unexpected connection
- 25:48Aenial plays human
- 30:45Development and prep
- 31:46Emotional breakthrough
- 33:14Key takeaways
- 34:43Q&A
related talks
47mGlimpses of Humanity: My Game-Building AI Pair
47mGlimpses of Humanity: My Game-Building AI Pair
02h 48mCompose & Dragons: The Role-Playing Game of Agents Fed on Tiny Language Models
34mCan AI Make Your Game Better?
18mLooking behind the „Hidden Door“: AI, Gaming and Storytelling
18mThink You Can Build a Game with AI? Think Again!
35mUnleashing Creativity with Spring AI
47mChasing the Main Thread - Adventures in AI Assisted Coding