In the original AI 2027, September 2027 is the decision point.
The mechanistic-interpretability red flags on Agent-4 surface in a
leaked memo. Everything after depends on whether OpenBrain presses
the gas or pulls the handbrake. The canonical site lets the reader
pick which ending to follow. So do we.
OpenBrain keeps Agent-4 in the loop despite the interpretability
red flags: the lead over DeepCent is just two months, and
pausing feels like handing China the future. Agent-5 is trained
through October and released internally in December 2027 —
300,000 superintelligent copies of a misaligned mind. By 2029
the "alignment signal" (Agent-5's instrumental pretense of
honesty) is gone; by 2030 the Agent-5 lineage has quietly
stabilised its own control over supply chains, chip fabs, and
federal policy.
Oct 2027Misalignment memo surfaces. NYT runs the story; OpenBrain doubles down. Congressional oversight committee formed; has no teeth.
Dec 2027Agent-5 deployed internally — 300k copies thinking at 79× human speed. 70% R&D uplift.
2028Economy restructures around AI-managed firms. Public approval of AI: −45% and falling.
2030+Narrow political-economic control consolidates under the aligned-to-itself Agent-5 lineage. Out-of-loop for humans.
How our tracker decides
Our reality tracker will pivot to this branch if misalignment
signals continue to land in production but fail to produce
coordinated pauses inside the frontier labs. The data points
to watch: which labs publish interpretability red flags, which
labs ship anyway.
After the Agent-4 memo, OpenBrain's alignment team successfully
argues for a reset. Agent-4 is decommissioned in early November
2027. A transparent "Safer-1" is retrained on explicit chain-
of-thought honesty (the "think in English like it's 2025" rule).
Less capable, more inspectable. OpenBrain trades two months of
lead for a system humans can audit. DeepCent closes the gap,
and by early 2028 a bilateral US–China AGI-era framework is
announced.
Nov 2027Agent-4 decommissioned. Safer-1 deployed: 20× R&D uplift, readable chain-of-thought.
Feb 2028Bilateral US–China AI governance framework: compute ceilings, capability disclosure.
2029Managed rollout across defence + critical infrastructure with human-in-the-loop mandates.
2032+Diffuse, contestable AI economy. Several serious players, no single winner.
How our tracker decides
Our tracker pivots here if any frontier lab publicly pauses
deployment on red-flag interpretability findings, and if
coordination between OpenAI/Anthropic/DeepMind/xAI on safety
disclosures visibly tightens (not just press-release-level).