This scenario describes the emergence of Artificial General Intelligence (AGI) — represented by Agent-3 — an intelligence equal to that of a human, followed by the rise of Superintelligence (Agent-4), far surpassing human intelligence, around 2027. There are two possible outcomes: ‘race’ — where the AI goes rogue (misaligned), or ‘slowdown’ — where security measures keep the AI under control (aligned). Competition with China adds complexity, and the window to make a decision is narrow.

https://ai-2027.com/