AI summary by TwinMind

AI-augmented game development with Inworld co-founder Kylan Gibbs

Sequoia Capital

7m 18s

Inworld AI Platform Overview

  • End-to-end vertically integrated platform specifically for gaming
  • Three core components:
    • Engine: drives experiences at runtime, connects to Unreal/Unity, takes game state input
    • Studio: crafts and augments existing game development using AI
    • Core solution: addresses gaming challenges (local execution, no latency requirements)
  • Game developers are technically savvy and creatively demanding, want control

Recent Launches and Partnerships

  • This week launched Project Explora with Nvidia partnership
  • Microsoft Studios collaboration
  • Ubisoft experience demonstration
  • Early experience game releases

Gaming AI Implementation Challenges

  • Games require local execution with no latency
  • Controls must align with narrative designers and writers' vision
  • Need gamified experience, not just chatbot functionality
  • Multiple models running in parallel for voice, multimodal input processing
  • Large models don't work effectively for parallel processing requirements

Demo Features Showcased

Conversational Dynamics

  • Enhanced dialog with game design elements
  • Character relationship progression mechanics
  • Emotional state updates in backend
  • Resistance recruitment conversation with burger preference discussion
  • Character personality consistency (Bloom's revolutionary passion vs Nelson's casual approach)

Dynamic Gameplay Mechanics

  • Real-time game state changes triggered by AI
  • UI elements dynamically modified based on conversations
  • Recon mission with drone control
  • Guard detection and stealth mechanics
  • Character height inquiry (6 feet response)

Quest Generation System

  • Dynamic quest creation on-the-fly
  • Generated based on player interactions
  • Grounded in established player goals
  • Mission planning discussions (gallery window entry, ladder vs grappling hook options)
  • Adaptive strategy selection based on player suggestions

Technical Architecture

  • Series of parallel models processing inputs
  • Signals sent to ultimate prompt system
  • Integration of voice, multimodal input, game state
  • Real-time character behavior adaptation

TwinMind

Get summaries like this for your own meetings

TwinMind takes notes in the background during meetings, lectures, and calls, then writes the summary for you.

Free plan to get started

No bot joins your calls

Phone, desktop & Chrome apps

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome