1) Four CIO-style choices per dilemma
Each card now offers four distinct approaches that map to recognizable leadership styles: Operator, Builder, Strategist and Guardian. This creates realistic trade-offs instead of one obvious optimal line.
This page summarizes the latest major updates using the full v4 release notes as source material. The short version: the game now behaves more like real-world AI transformation, where style, timing, trust and hidden costs matter as much as raw speed.
Each card now offers four distinct approaches that map to recognizable leadership styles: Operator, Builder, Strategist and Guardian. This creates realistic trade-offs instead of one obvious optimal line.
Win outcomes are now style-based and split by conduct quality. Your ending is tied to your actual decision pattern across the run, not just a simple meter snapshot.
Sub-options no longer show visible numeric impacts before you commit. Every option carries upside and downside, so you must reason from context and consequences.
Even successful runs now surface hidden costs: weakened trust, thin controls, debt buildup, or stalled transformation momentum. Surviving is separated from truly delivering.
New deployment can depress value before compounding gains appear, matching real transformation lag.
Adoption follows an S-curve and is gated by trust, capability and morale rather than executive mandate alone.
Trust now builds slowly and breaks quickly; incident and conduct penalties are sharper than positive recovery.
Backlog creates recurring drag; incident risk scales with control gaps and operational complexity.
The competitive baseline keeps moving, while low morale can trigger capability loss and cascading exits.
Board expectations rise when you perform, so maintaining credibility requires sustained delivery, not one strong quarter.
Runs feel less puzzle-like and more like executive leadership under uncertainty. You are rewarded for coherent strategy and punished for unmanaged second-order effects.
Debates become richer because each option is plausible from a different leadership posture. The discussion shifts from "right answer" to risk appetite and governance maturity.
For the full technical rationale, evidence links and balancing notes, read RELEASE-NOTES-v4.md.
Brynjolfsson et al., The Productivity J-Curve (AEJ Macro) and the productivity paradox literature.
EU AI Act, NIST AI RMF, DORA, and SR 11-7.
Knight Capital SEC order, Normal Accidents, and OWASP Top 10 for LLM applications.